JPS598878B2 - Automatic detection method of speed distribution of automobile traffic flow using photoelectric conversion elements - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for automatically detecting the traveling speed distribution in a vehicle traffic flow using a plurality of photoelectric conversion elements and a simple electronic circuit for the purpose of detecting abnormalities in the vehicle traffic flow. be.

BACKGROUND ART Conventionally, two or more photoelectric conversion elements are used as a traffic flow sensor, and traffic flow is measured using the difference in time between vehicles passing through detection surfaces set on the road surface.

However, in this type of measurement method, traffic flow sensors are installed on the assumption that the car is driving correctly within the lane, so it is relatively accurate if the car's driving trajectory, driving position, etc. are normal and do not change unnecessarily. Although a detection operation is performed, in actual traffic flow, cars may sometimes cross lanes, so it is unavoidable that a considerable error will occur.

Therefore, in practice, a method is used to supplement the detection signal by appropriately processing the detection signal transmitted through the telephone line using software.

Additionally, a method is already known in which the speeds of vehicles traveling on the target road surface are measured and those speeds are calculated as an average value, but this method averages out unusual speeds and also eliminates some error components. It is difficult to extract and separate the error components.

In view of the above, the present invention measures the running speed distribution of automobile traffic flow based on the outputs of a plurality of photoelectric conversion elements whose detection surfaces are set on the target road surface, and is used for early detection of abnormalities in traffic flow, etc. , it is characterized by being able to accurately grasp temporal and spatial changes in the running speed distribution situation on the target road surface.

Additionally, the present invention is capable of measuring vehicle traffic flow, compressing it into a statistical quantity called vehicle speed distribution, and transmitting it electronically to a management center, etc. via a telephone line. It is characterized by being able to increase efficiency.

Next, the method of the present invention will be explained in more detail based on the drawings.

FIG. 1 shows the configuration of a device for detecting the speed distribution of motor vehicle traffic according to the present invention.

As shown in the figure, a lens system 3b forms detection surfaces 3a and 4a smaller than the size of the automobile 2 on the target road surface 1.
, 4b, when two photoelectric conversion elements 3 and 4 are provided,
In these photoelectric conversion elements 3 and 4, changes in output signals occur in accordance with changes in the brightness of a car passing over the target road surface 1.

In addition to these, changes in the output of the photoelectric conversion elements 3 and 4 occur due to changes in illuminance on the target road surface 1, shadows caused by other objects, and weather phenomena such as rain and snow. Consideration must be given to distinguishing disturbance phenomena from passing automobile phenomena, such as automatic aperture of the lens system that responds to changes in road surface illumination, and automatic compensation circuits added to the processing circuit. The detection surfaces 3a and 4a on the target road surface 1 are formed at desired intervals in the traveling direction of the vehicle 2. Therefore, when a vehicle travels on the target road surface 1 in the direction of the arrow, the vehicle enters first. A change in output signal occurs in the photoelectric conversion element 3 corresponding to the detection surface 3a corresponding to the change in the brightness of the vehicle, and a similar output signal is generated in the other photoelectric conversion element 4 after a required time corresponding to the traveling speed of the vehicle. changes occur.

These output signals generated by the photoelectric conversion elements 3 and 4 are amplified by amplifiers I and II, differentiated by differentiating circuits I and II, and then converted by comparators I and II to a level with a temporal rate of change close to zero. Convert to square wave signal.

The output signal of the comparator circuit (■) is generated by a large number of delay circuits 1, II, etc. using monostable multivibrator circuits, etc.
・After delaying the pulses by the time D1, D2, . . . respectively, the waveform shaping circuits I, I, . Apply.

In addition, the output signal of the other comparison circuit (■) is the waveform shaping circuit (X).
After converting the pulse width to a constant pulse width, the AND circuits I, I
,... is applied.

Here, the delay times D1, D2, . . . are values that are successively changed, depending on the distribution of time during which a vehicle traveling on the target road passes between the two detection surfaces 3a and 4a. In relation to the width of the pulse determined by the waveform shaping circuits I, .
A pulse signal is output from any one of II, .

In other words, for example, the speed distribution of a car traveling at an arbitrary speed is
..., 48~52kf1l/hr, 52~56kr11
/hr,... When detecting for each stepwise speed division,
In the delay circuit (■) corresponding to 48 to 52 km7'hr, the delay corresponds to the time difference between the output signals of comparison circuits I and II when the car travels at a speed of 50 krrv'hr as a typical value. In the delay circuit V, which is set to have a time D4 and similarly corresponds to 52 to 56 krrv'hr, the car
The vehicle is set to have a delay time D corresponding to the time difference between the output signals of the comparator circuits I and (2) when the vehicle runs at speed r.

Therefore, the number of delay circuits etc. is determined by the width of the speed distribution of the vehicle expected on the target road surface 1 and the width of the speed division.

In addition, the waveform shaping circuit I. I[,... and the time width of the pulses output by the waveform shaping circuit is set to
From this, the width of the speed division is determined by the time width of the pulse output by the waveform shaping circuit.

Therefore, if the car 2 passes through the detection surfaces 3a and 4a,
A pulse will be generated from one of the AND circuits.

In this way, when the vehicle 2 passing between the detection surfaces 3a and 4a of the target road surface 1 is detected, and a pulse is output from one of the AND circuits, this pulse is sent to the counters I and I at the next stage, respectively. , . . . is constantly counted until it is reset at regular intervals.

As a result, in the speed distribution determining section at the final stage, a speed distribution as shown in Figures 2 to 4 is obtained for each cycle of the reset pulse, and the local maximum value is determined based on this speed distribution. If there is one maximum value as shown in FIG. 2, the corresponding running speed ν4 and the state of dispersion around the maximum value, such as the standard deviation, are calculated and output.

Further, when there are two maximum values as shown in FIG. 3, the corresponding velocities ν3 and ν7, their frequency of occurrence, standard deviation around them, etc. are calculated and output.

Furthermore, if the local maximum value cannot be determined as shown in FIG. 4, the entire velocity distribution is output, and the average value is calculated as necessary.

These outputs from the speed distribution determining section are displayed on a printer as a diagram or table substantially similar to those in Figures 2 to 4, or sent to a control center via a telephone line to determine the running speed distribution of vehicles on the target road surface. Based on the temporal and spatial changes in traffic, it can be used for early detection of abnormalities in traffic flow due to accidents or illegal vehicles by automatic means.

The velocity distribution measurement based on the device shown in FIG. 1 can also be carried out by a software-dependent device using a microcomputer as shown in FIG.

That is, in FIG. 5, as in the case of FIG. 1, the lens system 13 forms the detection surfaces 13a and 14a on the target road surface.
The output signals of these photoelectric conversion elements 13 and 14 are amplified by amplifiers I and ■, differentiated by differentiating circuits 1 and II, and then sent to a comparison circuit. 1 and II, it is converted into a rectangular wave signal sequence, which is shaped into a pulse signal sequence with a constant pulse width by a waveform shaping circuit 1 and If, and then input to the interface I of the microcomputer.

This interface (2) is connected to a CPU (central processing unit) via a common bus, and the CPU searches for the presence of a pulse signal by a polling method according to a program stored in a ROM.

If a pulse signal from the waveform shaping circuits 1 and II is present, the time at that time is read from the time information generating section via the same interface, and sequentially stored in corresponding addresses in the RAM.

Here, the pulse width of the pulse signals from the waveform shaping circuits I and II is set to a value greater than one polling cycle for the interface and smaller than two cycles.

By setting the above pulse width in this way, when a pulse signal is generated, the microcomputer can detect the occurrence at least once, and in some cases, the same pulse signal can be detected twice in a row. However, the microcomputer performs processing to ignore the second detection.

In the microcomputer, the following calculation is performed for the generation time groups T1 and T2 of the pulse signal series stored in this way, and ΔT (travel time between the detection surfaces 13a and 14a) that satisfies the condition of this equation is calculated. ) is calculated.

Here, τ corresponds to the quantum width (velocity division) on the horizontal axis of the velocity distribution shown in FIGS. 2 to 4.

What is obtained by this calculation etc. is substantially the same as the case explained with reference to FIG. 1, and is outputted to a printer via the interface (2) or sent to the management center via a telephone line.

In this way, the traveling speed of the vehicle on the target road detected by the photoelectric conversion element can be determined not as a fixed value but as a type of probability distribution.

As a result, it is possible that error components caused by causes other than the vehicle's running speed can be significantly eliminated by setting a determination program unique to each measurement point.

In the above, we have explained the method of setting two detection surfaces on the target road surface using two photoelectric conversion elements and measuring the traveling speed distribution at those points. It can be used as a method for early detection of abnormal phenomena occurring in traffic flows such as.

That is, since traffic signals generally do not exist on tunnels or bridges, it is assumed that the traffic flow thereon has a substantially uniform speed distribution over the entire section.

Therefore, by measuring the above-described traveling speed distribution over the entire tunnel or bridge, it is possible to automatically detect abnormalities in the traffic flow over the entire section.

FIG. 6 shows the configuration of a group of microphotoelectric conversion elements 20 used for such a purpose, in which microphotoelectric conversion elements 21, 21, .
... are arranged.

Such a microphotoelectric conversion element group 20 can be easily obtained by processing a silicon wafer using a photo-etching method.

As shown in FIG. 7, the micro photoelectric conversion element 20 is placed on the target road surface 22 as in the case of FIG. 1, and a road image is formed on the photocathode by the lens system 20a. A large number of detection surfaces 21a, 21a, . . . are located on the road surface 22.
form into an array.

The output signals of these minute photoelectric conversion elements 21, 21, . . . are sent to amplifiers I, n, .
... and differentiated by differentiating circuits I, II, ..., converting it into a rectangular wave signal sequence in comparison circuits I, II, ..., and converting it into a rectangular wave signal sequence by waveform shaping circuits I, I[, ... After being shaped into a pulse signal sequence with a constant pulse width, the pulse signal is input to the microcomputer via the interface I using a polling method.

This microcomputer performs processing similar to that described with reference to FIG. 5 on inputs from mutually adjacent minute photoelectric conversion elements, and calculates the running speed distribution at multiple points on the target road surface at once.

From this calculation result, it is determined that there is steady traffic flow in areas where the traveling speed distribution at each point on the target road surface is roughly similar.
If the distribution of a part of the road changes significantly due to some reason, it is immediately determined that an abnormality has occurred, but roads generally have individual characteristics depending on their environment, and the criteria for abnormality determination differ depending on the target road surface. Therefore, it is necessary to set the criteria for the above abnormality determination by fully considering the normal traffic flow situation on the target road surface.

In addition, in the photoelectric conversion element group 20, it is necessary to set the distance between the detection surfaces corresponding to each photoelectric conversion element within several meters to maintain a strong correlation between adjacent photoelectric conversion elements. It is.

In this way, in the present invention, the traveling speed distribution of automobile traffic flow at the target point is automatically detected, and thereby the temporal and spatial changes in the true traveling speed can be measured. Compared to the case of measuring the average value of vehicle speed, it is possible to accurately grasp the actual state of vehicle traffic flow, and error components can be largely removed. It is also possible to automatically detect the occurrence of abnormalities in the automobile traffic flow on the road in the section.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an apparatus for detecting the speed distribution of automobile traffic flow based on the method of the present invention, and FIGS.
The figure is a graph illustrating the speed distribution detected by the above device, FIG. 5 is a block diagram showing different configuration examples of the device for detecting the speed distribution, FIG. FIG. 7 is a block diagram of a velocity distribution detection device using the above-mentioned microphotoelectric conversion element group. 1, 22...Target road surface, 2...Car, 3, 4, 13, 14...Photoelectric conversion element, 3a
, 4a, 13a, 14a... detection surface.

Claims (1)

[Claims] 1. Detection surfaces of a plurality of photoelectric conversion elements are set along a lane on a target road surface, and output signals of the photoelectric conversion elements are determined according to changes in the brightness of a car passing on these detection surfaces. After each is differentiated by a differentiating circuit, it is converted into a rectangular wave signal by a comparator circuit, and based on this rectangular wave signal, the vehicle speed corresponding to the temporal difference between the output signals of adjacent photoelectric conversion elements is appropriately stepped for each speed category. 1. A method for automatically detecting the speed distribution of automobile traffic flow using a group of photoelectric conversion elements, characterized in that the speed distribution of automobiles traveling on a target road surface is determined by detecting the speed distribution of automobiles traveling on a target road surface. 2 Using a micro-photoelectric conversion element group in which a large number of micro-photoelectric conversion elements are arranged in a perspective shape along the road lane, the detection surface of the large number of micro-photoelectric conversion elements is set along the lane on the target road surface. Then, the output signals of the photoelectric conversion elements corresponding to changes in the brightness of a car passing over these detection surfaces are differentiated by a differentiating circuit, and then converted into a rectangular wave signal by a comparator circuit, and based on this rectangular wave signal, Temporal space for detecting an abnormality in the flow of automobile traffic traveling on a target road surface by detecting vehicle speeds corresponding to temporal differences in output signals of mutually adjacent photoelectric conversion elements step by step for each appropriate speed category. A method for automatically detecting the speed distribution of automobile traffic flow using a group of photoelectric conversion elements, which is characterized by determining the speed distribution of a vehicle.