JPH0348399A - Traffic monitoring device and display device - Google Patents

Abstract

PURPOSE:To obtain accurate traffic information by executing fuzzy inference based on a detecting signal outputted from a vehicle detecting sensor arranged in the vicinity of a road and outputting a traffic state signal. CONSTITUTION:When a vehicle passes under an ultrasonic sensor 1, the sensor 1 senses the vehicle by the reflection of an ultrasonic wave from the vehicle and outputs a sense signal. Thereby, a pulse-like sense signal with short width is outputted at the rapid passage of the vehicle, and on the contrary a sense signal with long pulse width is outputted at the slow passage of the vehicle. Thus, a sense signal with the time width corresponding to the existence time of the vehicle in the sensing area is obtained, but the time width of the sense signal is the width corresponding to a value longer than the practical length of the vehicle by the sensing area. The shadowed part is the time width based on the sensing area and the part other than the shadowed part is the time width due to the existence of the vehicle. Thus, fuzzy inference based on the detecting signal from the sensor 1 is executed and a traffic state signal corresponding to the practical traffic state is outputted.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a traffic monitoring device that detects the degree of traffic congestion of vehicles passing through a motorway, and a display device used in the monitoring device.

(Prior art/problems to be solved by the invention) In conventional traffic flow detection, the number of vehicles passing per unit time and average speed obtained by vehicle detectors are set to a certain threshold.
The degree of congestion is calculated by discriminating by level, and the traffic congestion display and signal i! It was used to control X.

However, with this conventional method, situations such as those at the border between the threshold level sometimes occur that do not match the actual traffic congestion situation, and do not necessarily correspond to the actual traffic flow.

The present invention has been made in view of the above circumstances, and it is an object of the present invention to provide various traffic monitoring devices capable of detecting the degree of congestion, etc., close to the actual traffic flow.

Furthermore, in order to achieve such an object, the present invention according to claim (1) includes a table suitable for expressing the degree of traffic congestion and its changes. , and a fuzzy inference function that performs fuzzy inference based on the detection signal from the vehicle detection sensor and outputs a traffic condition signal.

According to the invention of claim (1), fuzzy inference is performed in the fuzzy inference section based on the detection signal from the vehicle detection sensor, and a traffic condition signal corresponding to the actual traffic condition is output.

In the invention of claim (2) of '51', the traffic monitoring device includes a vehicle sensor that detects a passing vehicle and outputs a signal indicating the passing of the vehicle, and a vehicle that processes the output signal of the vehicle sensor. A first arithmetic processing unit outputs a signal representing the average speed of passing vehicles and a signal representing the number of passing vehicles, and a membership function representing a congestion grade using the average speed of passing vehicles as a variable is set. , a first processing unit that outputs a corresponding membership function value in response to input of a signal representing an average speed given from the first arithmetic processing unit.
A membership function output means is set, and a membership function representing a congestion grade with the number of passing vehicles as a variable is set, and a membership function outputting means for outputting the number of passing vehicles is set in response to input of a signal representing the number of passing vehicles given from the first arithmetic processing unit in Section 11j. performing a fuzzy inference operation on the membership function values output from the second membership function output means for outputting the corresponding membership function b″i, and the first and second membership function output means; The vehicle is configured to include a second calculation means that outputs a signal representing the degree of traffic congestion.

According to the invention of claim (2), the average speed measured by using the membership function of /112 congestion degree with the average speed of passing vehicles as a variable and the membership function of congestion degree with the number of passing vehicles as a variable. Since the degree of congestion is calculated by applying the number of vehicles and the number of passing vehicles to these membership functions, it is possible to obtain a degree of congestion output that almost matches the actual traffic congestion.

In the invention of claim (3), the traffic monitoring device includes a vehicle sensor that detects a passing vehicle and outputs a signal indicating the passing of the vehicle, and a vehicle sensor that processes the output signal of the vehicle sensor 23 to detect the passing vehicle. a first arithmetic processing unit that outputs a signal representing a change in the average speed of passing vehicles and a signal representing a change in the number of passing vehicles; and a membership function representing a congestion grade using the average speed change of passing vehicles as a variable. , a first membership function output means for outputting a corresponding Mennokushi-Sobu performance value in response to input of a signal representing an average speed change given from the first arithmetic processing unit; A membership function representing a traffic congestion grade with a change in the number of passing vehicles as a variable is set, and the corresponding membership function value is calculated in response to the input of a signal representing a change in the number of passing vehicles given from the first arithmetic processing unit of 1111. The second man to output! ＜Performing fuzzy inference operation γ on the menharship score output means and the menharship score output means from the first and second menharship function output means to represent the trend of change in retention degree. The configuration includes a second calculation means that outputs a signal.

According to the invention of claim (3), the average measured by using the membership function of the degree of congestion whose variable is the change in the average speed of passing vehicles and the membership function of the degree of congestion whose variable is the change in the number of passing vehicles. By applying changes in speed and changes in the number of passing vehicles to these membership functions, it is possible to find out changes in the degree of congestion that correspond to the actual degree of congestion.

In the invention of claim (4) of j-i, the display device is configured to display the tortoise to be displayed in a direction in which the rate of change in the amount is projected or recessed from one end of the display box according to the length of the display band. The structure is expressed by the length of each part.

According to the invention of claim (4), there is (-H and the change in its amount in one representation Ij'I
? , : Ishi can be expressed using a bar graph display, so the display is easy to understand and the way it is displayed is simple!
F, and the display area can also be reduced.

In the invention of claim (5), the traffic monitoring device includes a plurality of vehicle detectors arranged along lanes of a motorway, and a vehicle that responds to a vehicle stagnation situation given from the plurality of vehicle detectors. and a vehicle stagnation status monitoring unit that operates in response to input of data regarding the sensing time length, and the vehicle stagnation status monitoring unit operates in response to input of data regarding the sensing time length. The configuration includes means for fuzzy inferring the state of stagnation of vehicles on the motorway according to fuzzy rules using data.

According to the invention of claim (5), the vehicle stagnation situation monitoring section includes:
Data regarding vehicle detection time length is inputted from a plurality of vehicle detectors arranged along the lanes of the motorway, and the vehicle stagnation status monitoring unit uses vehicle detectors 23 arranged adjacent to each other.
Fuzzy inference is made as to the state of vehicle stagnation on the motorway according to fuzzy rules using data related to vehicle sensing time lengths that are sequentially input from each of the above.

In the present invention according to claim (6), the traffic monitoring device includes a plurality of vehicle number detectors arranged along lanes of a motorway, and a vehicle stagnation situation given from the plurality of city vehicle number detectors. and a vehicle stagnation situation monitoring section that operates in response to input of data regarding the number of detected vehicles corresponding to the vehicle stagnation situation monitoring section, and the vehicle stagnation situation monitoring section operates in response to input of data regarding the number of vehicles detected corresponding to the number of vehicles detected. Using data regarding the number of vehicles, the fuzzy H
The structure is equipped with a means to discuss the following points.

According to the invention of ai'r requirement (6), data regarding the number of vehicles detected is input to the vehicle stagnation situation monitoring unit from a plurality of ilj vehicle number detectors arranged along the lanes of the motorway. Ru. The vehicle stagnation status monitoring unit performs fuzzy inference on the stagnation status of vehicles on the motorway according to fuzzy rules using data related to the number of detected vehicles that are sequentially input from each of the vehicle number detectors arranged adjacent to each other.

In the light of claim (7), the traffic monitoring device is equipped with two ultrasonic detection devices that detect the vehicle speed, vehicle interval, and number of passing vehicles based on the sonic Doppler effect, wherein the traffic monitoring device comprises: t/7 ttt of vehicle speed, vehicle distance, and number of vehicles detected by ultrasonic detector
are input, and these h'. ? The configuration includes a fuzzy inference means for fuzzy inferring the degree of congestion and traffic volume corresponding to the vehicle speed, vehicle interval, and number of vehicles based on the information and according to fuzzy rules.

According to the invention I of claim (7), each information of the vehicle speed, vehicle interval, and number of passing vehicles obtained by the ultrasonic detector based on the ultrasonic Doppler effect is input to the fuzzy inference means. . The fuzzy inference means fuzzy infers the degree of congestion and traffic volume corresponding to the vehicle speed, vehicle interval, and number of passing vehicles based on this information and according to fuzzy rules.

9 blank rows (5 rows of vehicles) 1 is placed above the road,
It is configured to transmit ultrasonic waves towards the road surface and receive the reflected tig waves. When a vehicle passes below the ultrasonic sensor 1, the sensor 1 detects the vehicle by the reflection of the ultrasonic waves from the car's side. Outputs an angry signal. Sensing signal (:
The output can be read while the vehicle is floating within the detection area of the detector 1. Therefore, when a vehicle passes at high speed, a short pulse-like sensing signal is output, and on the other hand, when a vehicle passes at low speed, a sensing signal with a long pulse width is output. In this way, a sensing signal with a time width corresponding to the presence time of Na'-I can be obtained in the sensing area of the vehicle. The time width of this detection signal is a width corresponding to a value corresponding to a value corresponding to the sensing area from the value of the vehicle in the dark. Second
The shaded area in the diagram is the time width due to this sensing area, and the area other than the shaded area is the time width due to the presence of a vehicle in the dark.

FIG. 3 is for explaining how to calculate the number of passing vehicles and the average speed from the sensing signals obtained by the vehicle sensor 1. This figure shows the sensing signals during off-peak hours and during busy times.

During some arbitrary time interval T. Suppose that two vehicles pass by. Since two vehicles passed by, the 4w% signal is output as two pulses, one during off-peak hours and one during congested traffic.

However, the time width of the sensing signal during off-peak hours is t11”1
2, it is a sufficiently small value for the period T, but
When the traffic becomes congested, the time width of the sensing signal becomes t21. It occupies most of the period T, as indicated by t22.

The number of passing vehicles is determined by counting the number of pulses of the sensing signal per unit time.

In determining the average speed of passing vehicles, the concept of average vehicle length is used. In general, many types of vehicles (trucks, cars, buses, etc.) pass on roads, and the actual length of each type of vehicle can be obtained by weighting it by the mix rate of each type of vehicle on the road. Let the average value of all vehicle types be the average vehicle length. Add the length of the sensing area of the vehicle sensor to the average vehicle length. Multiply this addition result by the number of passing vehicles per unit time. The average velocity is obtained by dividing the result of the two multiplications by the sum of the time widths of the sensing signals per unit time. That is, the average speed is given by the following equation.

Figure 4 shows the relationship between the average speed of passing vehicles and the traffic congestion grade. This is called the congestion grade membership function. The traffic congestion grade indicates the degree of traffic congestion; the higher the grade, the worse the traffic congestion. The graph in Figure 4 shows that when traffic congestion slows down the flow of vehicles, the average speed of each vehicle slows down.

Figure 5 shows the relationship between the number of passing vehicles and the traffic congestion grade. Each road has its own appropriate number of vehicles to pass through. If there is no traffic jam, the number of vehicles passing will be around this appropriate value. As the traffic congestion increases, the number of passing vehicles may decrease or increase depending on the case.

The membership function of the degree of change in traffic congestion (change in traffic congestion grade), which indicates whether traffic congestion is about to start or is on the way to being resolved, is shown in FIGS. 6 and 7. Figure 6 shows the relationship between the average speed change and the degree of change in traffic congestion (degree of increase in traffic congestion, degree of decrease in traffic congestion), and Figure 7 shows the relationship between the change in the number of passing vehicles and the degree of change in traffic congestion.

Figure 6 is. When the change in average speed increases in the positive direction, congestion tends to decrease (congestion is eliminated). If the train moves too far to the tiger side, it indicates that the traffic is heading in the opposite direction. Figure 7 shows that as the change in the number of passing vehicles increases in the positive or negative direction, the amount of change in traffic congestion increases.
(The average speed and the number of passing vehicles are determined by the method described above. A digital signal representing these is output. Also, in the arithmetic unit 2.
The changes (deviation, differential value) in the average speed and number of passing vehicles per unit time are determined. Digital signals representing changes in average speed and changes in the number of passing vehicles are output.

on the other hand. Four memories or memory areas 4, 5, 6, and 7 are provided. These memories include sections 4.5 and 6.7.
The membership functions shown in the figure are stored in advance. The horizontal axis of these membership functions,
That is, the average speed, number of passing vehicles, change in average speed, and change in number of passing vehicles correspond to the address. Then, the average speed and number of passing vehicles are output from the computing device rt2. By specifying the addresses of these memories 4, 5, and 6.7 by digital signals representing changes in average speed and changes in the number of passing vehicles, the S traffic congestion grade corresponding to that address is determined.
Data representing the degree of change in traffic congestion is read out.

The data representing the traffic congestion grade read from the memories 4 and 5 is applied to the MIN circuit 8, whichever is smaller is selected and a signal SH representing the selected data is output. Similarly, the data representing the degree of change in traffic congestion read out from memories 6 and 7 is given to the MIN circuit 9, which has the smaller absolute value and is the same as the data output from memory 6. signal S with sign (positive or negative),
is output from the MIN circuit 9. These signals sH,S
h is given to the display device 10.

The display device rIt10 is a display device capable of displaying image information, and expresses the degree of traffic congestion and its changing tendency through images as shown in the block of the display device 10 in FIG. 8 or as shown in FIG. .

That is, this image is composed of a bar graph-shaped strip and a direction display section that is triangularly protruded (FIG. 8) or recessed (FIG. 9) from the upper end of this strip. The height H of the band-shaped portion expresses the degree of congestion represented by the output signal Su of the M I N circuit 8. The higher the height H, the more severe the traffic congestion. Even if this height H is below a certain level (f there is no traffic jam), the direction display section indicates whether the traffic jam is increasing or decreasing, and also indicates the degree.The output signal Sh of the MIN circuit 9 Even if is positive, the 1-bit traffic jam tends to increase, so it forms a protruding triangular shape (Fig. 8), and its height h indicates the degree of change in traffic congestion represented by the signal Sh. If S, is negative, the traffic jam is decreasing, so it becomes a concave shape (Fig. 9).The height h indicates the degree of traffic congestion change caused by the signal Sh.The direction display section is It can be not only a triangle but also other shapes (arrow, arc, trapezoid, etc.).

As above. In the display 'It H L O,
The height of the graph represents the degree of congestion. The direction and size of the arrow will represent the direction and degree of increase or decrease in traffic congestion. Not only the display device but also other output devices can be used to output the congestion degree and its degree of change.

Shizu T Sen 1-J Hereinafter, other tip embodiments will be explained.

Figure 10 is a diagram showing the configuration of the traffic monitoring system (traffic monitoring device) for monitoring vehicle stagnation on expressways.
Figure C is a diagram showing a fuzzy logic 111 provided in the vehicle stop in Figure 10>It situation monitoring unit;

The vehicle stagnation situation monitoring system of this embodiment is based on the driving lane I1.
Data on vehicle detection time lengths that are arranged at regular intervals along the sides of lane rt of a road for automobiles 1l having an overtaking lane rtb and an overtaking lane rtb, and that change depending on the state of vehicle stagnation on that motorway. A plurality of vehicle sensors (vehicle detection sensors) I) 1,..., Dn,...
. . (generally referred to as l)i), and a vehicle stagnation status monitor 'l'i' which operates in response to a data input regarding the vehicle sensing time length from each vehicle sensor D1.

Here, Ct+1 is a stalled vehicle, Ca is a vehicle traveling in the driving lane 11a, cb is a vehicle traveling in the passing lane nb, and when the stalled vehicle CI is present at the position shown in the figure. When the following vehicles Ra, Ilb drive in the lane along the vehicle sensor installed near the stopped vehicle Ct, their running speed is reduced because their running is obstructed by the stopped vehicle Ct. As a result, vehicle detectors near the stop vehicle Ct will output data with a vehicle detection time length longer than the vehicle detection time length output from other vehicle detectors. will be done. Then, using the above-mentioned fact that the vehicle detection time length changes depending on the vehicle stagnation situation, the following fuzzy ti
By using T-theory, it is possible to grasp the situation of vehicle stagnation by identifying the stagnation point of the vehicle.

Therefore, the vehicle stagnation situation monitor T T l;l, 11th
As shown in the figure, the fuzzy inference means includes an input device IN, a fuzzy inference section FS, a fuzzy rule storage section FM, and an output device OUT.

The input device IN is constituted by a plurality of sets of at least three of the plurality of vehicle sensing devices Di arranged adjacent to each other, and three sets of the I-th and I-th devices forming the I II1. 2. Third vehicle sensing P, H)n, Dn + l,
Vehicle sensing time length data Dn from each of Dr+ +2
, Dn +l, Dn +2 (where n is a positive integer)
are individually output to the fuzzy inference FS as antecedent variables for sequentially performing fuzzy inference.

The fuzzy rule storage unit FM stores if as shown in FIG.
It stores a plurality of types of fuzzy rules, 22 types in this example, in the format (antecedent part) to then (consequent part).

In each fuzzy rule shown in FIG.
, Dn +1, Dn +2 are the corresponding second and third vehicle sensors Dn, Dn +1, Dn +2, respectively.
Antecedent variable Y t,y regarding vehicle sensing time length data given from 2. Y3 is each vehicle detector Dn
, Dn + I, and Dn + 2 are installed consequent variables, PS,
PM and PL are fuzzy label names of fuzzy sets to which the antecedent and consequent variables belong, respectively, where PS is a fuzzy label name indicating positive small, PM is positive medium, and PL is positive large. .

The fuzzy inference part FS uses antecedent variables Dn, Dn+l, Dn+ as shown in FIGS. 13(a), (b), and (c).
2. Membership functions in each membership function coordinate system are stored. Figure 13(a) is the membership function for the antecedent variable Dn, Figure 13(b)
shows the membership function for the antecedent variable Dn+1, and FIG. 13(C) shows the membership function for the antecedent variable Dn+2. In addition, the whistle l3 figure (a) (b)
In (c), the numerical values of the front P part variables shown on the horizontal axis indicate the length of r8 at the time of vehicle detection, and PS, P
- M and PL each correspond to the membership functions illustrated below.

Fuzzy reasoning part 1? S is also shown in Figure 14 (a) (b) (
C), the consequent variable yl. Membership functions in the membership function coordinate systems of y2 and y3 are stored. Figure 14(a) shows the consequent variable y1
14(b) is the membership function for the consequent variance y2, FIG. 14(C)
respectively indicate member nop functions for the consequent variable y3. Also, Figure 14 (a) (b)
In (c.), the consequent variables shown in the axis test indicate the stagnation status of the vehicle by ``possible stagnation,'' ``almost stagnation,'' and ``definitely stagnation,'' and PS, PM, and PL are shown below. Membership functions are supported.

The fuzzy inference result Dn. from such fuzzy inference unit FS. Dn + 1' and Dn + 2' are output to the respective corresponding output devices OUT.

The operation of the fuzzy inference unit PS will be explained.

The vehicle sensing time length given from the first vehicle sensor Dn, that is, the antecedent variable Dn, the vehicle sensing time length given from the second vehicle sensor Dn+1, that is, the antecedent variable Dn+I, and the third vehicle sensor Dn+2. Figure 13 (
Membership values that fit the corresponding membership functions of each fuzzy rule are determined from a), (b), and (c), respectively.

Then, for each fuzzy rule, each previous (tf− +El
! The smaller membership value of DnDn+IDn+2 is selected (MIN operation), and the selected membership value determines the y l. of each fuzzy rule from FIGS. y2. The membership functions of PS, PM, and PL for each of y3 are cut. PS for each of y I, y 2, y 3 of all these cut fuzzy rules,
PM, 1) Each membership function of L is superimposed (MAX operation) to obtain the final y l. y2. A superposition membership function for each of y3 is obtained. By determining, for example, the center of gravity of this superposition membership function, data regarding the determined vehicle stagnation situation can be obtained.

In the above embodiment, the vehicle stagnation situation is inferred based on the vehicle detection time length data obtained from the vehicle detection 22, but when there is a vehicle stoppage FIB, the number of vehicles passing through the stagnation point also depends on the vehicle stagnation point. Since it changes depending on the situation, it is possible to use a vehicle number detector instead of a vehicle sensor as a means of understanding the vehicle stagnation situation. Therefore, in another embodiment of the present invention, the vehicle number detector 2e is installed at regular intervals along the side of the lane of the motorway, and the detected number of vehicles output from the vehicle number detector is used in the embodiment described above. Similarly, data regarding the number of vehicles detected from each vehicle number detector is used as data regarding the stagnant state of vehicles and is set to the antecedent variables On, On +1.
On +2, and each antecedent variable On
, On + 1, and On +2 are given to the fuzzy inference unit FS, respectively. Then, the fuzzy rule storage unit FM stores fuzzy rules as shown in FIG.
, On +2, each membership function is expressed as the 16th
The settings as shown in Figures (a), (b), and (c) are stored in the fuzzy inference FS, and the membership function and each of the antecedent variables On, On + 1, On +2 are used to It is also possible to make fuzzy inferences about the stagnation situation. The numerical values on the horizontal axis in FIGS. 16(a), (b), and (c) are the numerical values of the antecedent variable indicating the number of detected vehicles, and the consequent variance in this case y I, y 2, The membership functions for y3 are the same as in FIGS. 14(a), (b), and (c), respectively.

With the above configuration, the vehicle stagnation situation monitoring system in this embodiment can detect the stagnation situation of vehicles such as stagnation points of stagnant vehicles on motorways by using the vehicle detection time length and the number of detected vehicles from the vehicle detection 2x. Since this is done using data and fuzzy inference, there is no need to rely on reports from vehicle drivers near the stagnation point or monitoring information from TV cameras placed near the stagnation point.

In addition, in the monitoring system of this embodiment, since the system does not receive a report, it becomes easy to detect and respond to vehicle stagnation situations such as traffic jams caused by stagnant vehicles at an early stage. in this case,
It is not necessary to install a large number of vehicle detectors so that the interval between them is shorter than the length of the vehicle, or to install a TV camera or the like to identify the stagnant point of a stagnant vehicle. The cost can be lower than before.

FIG. I7 is a block diagram of a traffic monitoring device according to yet another embodiment. The traffic monitoring device l5 of this embodiment is based on the ultrasonic Doppler effect, and the vehicle speed (k
m/hour), vehicle interval (m), and number of passing vehicles (vehicles/minute)
The ultrasonic detection sensor 20 that detects vehicle speed and the vehicle speed and vehicle interval ti'f information detected by the ultrasonic detector 20 are input, and the traffic light is activated based on this information. 40, a controller 30 that controls a display 50, etc. The controller 30 is equipped with a fuzzy inference means 60 that fuzzy infers the degree of congestion and traffic volume corresponding to vehicle speed, vehicle interval, and number of passing vehicles according to fuzzy rules.

The above ultrasonic detector 20, as shown in FIG.
Ultrasonic transducer 2 installed at an angle above the road surface L
0a, and a main unit 20b that performs data processing based on the detection signal of the ultrasonic transducer 3i; While transmitting and receiving sound waves, the main unit 20b calculates the vehicle speed, the distance between vehicles, and the number of passing vehicles based on the detection signal from the ultrasonic transducer 20a.

Further, the fuzzy inference means 60, as shown in FIG. 18, includes an input section IN, a fuzzy inference section PR, a Fanoy rule storage section PM, and an output section OUT. Input part I
N is each information signal S of the vehicle speed, vehicle interval, and number of passing vehicles calculated by the main body 20b of the ultrasonic detector 20.
Antecedent variable XI for performing fuzzy theory on S,,s3
N Xt and X, respectively, are inputted and given to the next stage fuzzy inference unit PR. Fuzzy rule storage section F'
As shown in FIG. 21, M is H (antecedent part) ~ then
It stores multiple types of fuzzy rules in the (consequent part) format. In each of these fuzzy rules, in the figure, the symbol ■ indicates each condition when a vehicle is passing on the road, ■ indicates each condition when there are no passing vehicles on the road, and ■ indicates each condition due to traffic congestion. These are the conditions when all vehicles are stopped. Further, κIs F, X, respectively correspond to the vehicle speed, vehicle interval, and number of passing vehicles input to the input section IN. The antecedent variables y1, y, are the consequent variables of the congestion degree and traffic ■ that should be output as a result of fuzzy inference, and ZR. P.S., P.M.
, PL are the fuzzy label names of the fuzzy sets to which the antecedent variable X, , Xt and the consequent variables y1 and y, respectively, ZR is zero, PS is small positive, PM is positive medium, and PL is positive. The fuzzy label name indicating the size of each is shown.

Furthermore, the fuzzy inference unit FS is shown in FIGS. 22(a) and (b).
, (C>) Membership functions corresponding to the antecedent variables x1, Each membership function corresponding to the consequent variables Yr and Yx in the functional coordinate system is stored.In addition, the numerical value of the antecedent variable x1 shown on the drawing axis in FIG. 22(a) is the vehicle speed. (km/hour), the antecedent variable X shown on the horizontal axis in Figure (b)
, is the vehicle interval (m), the value of the antecedent variable It corresponds to the membership function illustrated below. Furthermore, in FIGS. 23(a) and 23(b), the consequent variable Y. The value of Yt indicates the degree of congestion and traffic volume with a value from O to 3, ZllSPS,
PM and PL respectively correspond to the membership functions illustrated below.

Next, a description will be given of the operation of determining the traffic congestion level and the traffic speed by the traffic monitoring device I5 having the above configuration.

When the vehicle 70 travels on the road L, the ultrasonic detector 20 detects the vehicle speed, the distance between vehicles, and the number of vehicles passing by as the vehicle 70 passes. Of course, since the Doppler shift of the ultrasonic wave varies depending on the magnitude of the vehicle speed, the vehicle speed can be detected from the amount of this Doppler shift. In addition, as shown in Fig. 20, the interval 'r of the vehicle detection pulse t varies depending on whether the traffic is quiet ((a) in the figure) or when it is crowded ((b) in the figure).
1 and T2 change, so the vehicle interval can be known depending on the length of the intervals T1 and T2. Furthermore, the number of passing vehicles is measured by counting the number of detection pulses per unit time. And ultrasonic detection 'd'j2
From 0, each information signal s,, s. .
,X, is added to the next stage fuzzy reasoning F1.

The fuzzy inference F'R is a fuzzy label set in the antecedent {if) of each fuzzy rule in FIG. 21 based on the antecedent variance X1, κ, κ, given from the input part IN The named membership functions are shown in Figure 22(a), (
Select from b) and (C) and calculate the membership value (;fi degree) in their membership function.
seek. Then, for each fuzzy rule, the minimum membership value (fitness) of each antecedent part is selected (MI
N operations). Next, the membership function with the fuzzy label name set in the consequent part (then) for each fuzzy rule in FIG. 21 is shown in FIGS. 23(a) and (b).
The corresponding consequent part (the
Cut the membership function of n) horizontally. and,
ZR, PS, P of each of these cut fuzzy rules
The membership functions of M and PL are all superimposed (MAX operation), and finally the superposition membership functions regarding the consequent variables )'+ and Yt are obtained. By determining, for example, the center of gravity of this superimposed membership function, data y, , y, regarding determined vehicle speed, vehicle interval, and number of passing vehicles can be obtained.

The data y1 and 'It regarding the vehicle distance, vehicle speed, and number of passing vehicles are outputted to the traffic light 40, the display 50, and a central processing unit (not shown) via the output section OUT.

In this case, when the road is empty and there are no arrested vehicles, both the vehicle speed and the number of vehicles are zero, so the antecedent is x+=ZR, X3=ZR, and the consequent in this case is all y+ Since the membership function =ZH is selected, it is determined that there are no passing vehicles. On the other hand, if the road is congested and the vehicle is stopped, even if the vehicle speed is zero, the presence of the vehicle will cause the ultrasonic echo to be detected by the ultrasonic transducer 20a.
The number of vehicles will not be zero because the waves will be received at Therefore, the antecedent part is! ．． =Z RSxs=PS, and since the membership function of y+=PL is selected for all consequents in this case, it is determined that the road is congested.

According to the traffic monitoring device of this embodiment, the degree of washing and the degree of traffic filtration are determined using fuzzy inference based on the vehicle speed, vehicle interval, and number of passing vehicles obtained by ultrasonic detection 2. This makes it possible to reliably determine whether the vehicle is stopped due to traffic congestion or whether there are actually no passing vehicles.

(Effects of the Invention) As is clear from the above explanation, according to each of the traffic monitoring devices of the present invention, fuzzy inference is performed in the fuzzy logic section based on the detection signal from the vehicle extrusion sensor. As a result, a traffic condition signal corresponding to the actual traffic condition is output, making it possible to obtain accurate traffic information.

Furthermore, a display device that is easy to understand, easy to display, and requires only a small display area can be obtained, which is most suitable for a traffic monitoring device.

[Brief explanation of drawings]

Figures 1 to 9 relate to the first embodiment, where Figure 1 is a vehicle sensor arrangement diagram, Figure 2 is a vehicle detection signal diagram, and Figure 3 is a calculation method for the number of passing vehicles and average speed. Fig. 4 is a diagram showing the membership function of congestion grade against average speed, Fig. 5 is a diagram showing membership function of congestion grade against the number of passing vehicles, and Fig. 6 is a diagram showing the membership function of congestion grade against average speed. FIG. 7 is a diagram showing the membership function of the degree of change in traffic congestion with respect to changes in speed; FIG.
The figure is a block diagram showing the configuration of the traffic jam detection device, and FIG. 9 is a diagram showing another display example. FIGS. 10 to 16 relate to the second embodiment, FIG. 12 is a diagram showing the fuzzy rules stored in the fuzzy rule storage unit, and FIGS. 13(a), (b), and (c) are the membership functions in the antecedent variables, respectively. Figure 14 (a
), (b), and (C) are diagrams showing membership functions for consequent variables, respectively; FIG. 15 is a diagram showing fuzzy rules stored in the fun rule storage unit in another embodiment; and FIG. 16 ( FIGS. 8A, 7B, and 7C are diagrams showing membership functions corresponding to each antecedent variable in another embodiment. 17 to 23 relate to the third embodiment, FIG. 17 is a block diagram showing the configuration of the traffic monitoring device, FIG. 18 is a block diagram showing the fuzzy inference means of FIG. 17, and FIG. Explanatory diagram when detecting a vehicle with an ultrasonic detector, 2nd
Figure 0 is a time chart of the detection signal when detecting the distance between vehicles using an ultrasonic detector, and Figure 21 is a time chart of the detection signal recorded in the fuzzy rule storage unit. Diagram 22 showing fuzzy rules
Figures (a), (b), and (C) are diagrams showing the membership functions for the antecedent variables, respectively, and Figure 23 (a)
) and (b) are diagrams showing the membership functions in the posterior genital variables, respectively. 1, DI=Dn,...vehicle sensor (vehicle detection sensor), 20... ultrasonic detector (vehicle detection sensor),
I's...Fuzzy Reasoning Department.

Claims (7)

[Claims] (1) A traffic monitoring device comprising: a vehicle detection sensor disposed close to a road; and a fuzzy inference unit that performs fuzzy inference based on a detection signal from the vehicle detection sensor and outputs a traffic condition signal. (2) A vehicle sensor that detects a passing vehicle and outputs a signal indicating the passing of the vehicle; and a vehicle that processes the output signal of the vehicle sensor to produce a signal indicating the average speed of the passing vehicle. A first arithmetic processing unit that outputs a signal representing the number of passing vehicles, and a membership function representing a congestion grade using the average speed of passing vehicles as a variable are set, and the membership function represents the average speed given from the first arithmetic processing unit. A first membership function output means for outputting a membership function value corresponding to the signal input in response to the signal input, and a membership function representing a traffic congestion grade using the number of passing vehicles as a variable are set, and the first a second membership function output means for outputting a membership function value corresponding thereto in response to input of a signal representing the number of passing vehicles given from an arithmetic processing unit; and from the first and second membership function output means. A traffic monitoring device comprising: second calculation means for performing fuzzy inference calculation on the output membership function value and outputting a signal representing the degree of congestion. (3) a vehicle sensor that detects a passing vehicle and outputs a signal representing the passing of the vehicle; and a vehicle sensor that processes the output signal of the vehicle sensor to produce a signal representing an average speed change of the passing vehicle. A first arithmetic processing unit that outputs a signal representing a change in the number of passing vehicles, and a membership function representing a congestion grade using a change in the average speed of passing vehicles as a variable are set, and the average given from the first arithmetic processing unit is set. A first membership function output means for outputting a membership function value corresponding to the input of a signal representing a speed change, and a membership function representing a traffic congestion grade using a change in the number of passing vehicles as a variable are set. , a second processing unit that outputs a corresponding membership function value in response to input of a signal representing a change in the number of passing vehicles given from the first processing unit;
membership function output means; and a second member for performing fuzzy inference calculations on the membership function values output from the first and second membership function output means and outputting a signal representing a tendency of change in traffic congestion level.
A traffic monitoring device equipped with computing means and. (4) A display device in which the amount to be displayed is expressed by the length of a display band, and the rate of change in the amount is expressed by the length of a directional display portion that projects or is recessed in a tapered manner from one end of the display band. (5) A plurality of vehicle detectors arranged along the lanes of the motorway, and a vehicle stagnation that operates in response to input of data regarding the vehicle detection time length corresponding to the vehicle stagnation situation given by the plurality of vehicle detectors. The vehicle stagnation status monitoring unit detects vehicles on the motorway according to fuzzy rules using data regarding vehicle detection time lengths that are sequentially input from vehicle detectors arranged adjacent to each other. 1. A traffic monitoring device for a motorway, characterized in that it is equipped with means for fuzzy inference of stagnation conditions. (6) A plurality of vehicle number detectors arranged along lanes of a motorway, and a vehicle that operates in response to input of data regarding the number of vehicles detected corresponding to a vehicle stagnation situation given from the plurality of vehicle number detectors. and a stagnation status monitoring unit, the vehicle stagnation status monitoring unit is configured to detect vehicle stagnation status on the motorway according to fuzzy rules using data related to the number of detected vehicles that are sequentially input from each of the vehicle number detectors arranged adjacent to each other. 1. A traffic monitoring device for a motorway, characterized by comprising means for fuzzy inference of vehicle stagnation conditions. (7) In a traffic monitoring device equipped with an ultrasonic detector that detects the vehicle speed, vehicle interval, and number of passing vehicles based on the ultrasonic Doppler effect, the vehicle speed detected by the ultrasonic detector; The present invention includes a fuzzy inference means for inputting vehicle spacing and vehicle number information and fuzzy inferring the degree of congestion and traffic volume corresponding to the vehicle speed, vehicle spacing, and number of vehicles according to fuzzy rules based on these information. Characteristic traffic monitoring device.