JP2007073058A - Vehicle sensing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a vehicle sensing device capable of detecting with high accuracy a vehicle speed existent in a monitoring area on a road, and a vehicle sensing system having the vehicle sensing device. <P>SOLUTION: The vehicle sensing device detects a vehicle speed existent in the monitoring area on a road 100, and includes a first thermopile element and a second thermopile element each for sensing infrared emission from a sensing object, and a speed detection section for calculating the speed of a vehicle 200, using input level values obtained from both elements. Both elements are disposed so that each monitoring area 3A, 3B based on each element is arranged in the vehicle lane direction of one vehicle lane. The speed detection section includes a time difference calculation means for calculating the time difference of output waveforms of input level values obtained from respective elements, and a speed calculation means for calculating the speed of the vehicle 200 from the obtained time difference and the distance between the two monitoring areas 3A, 3B. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a vehicle detector that detects the speed of a vehicle, and a vehicle detection system including the vehicle detector. In particular, the present invention relates to a vehicle detector that can determine the speed of the vehicle more accurately, and a vehicle detection system that includes the detector.

  2. Description of the Related Art Conventionally, an ultrasonic detector or the like is well known as a vehicle detector that detects a vehicle in order to examine traffic flow such as traffic volume and occupation rate.

  In general, an ultrasonic detector uses a so-called active sensor that emits an ultrasonic wave by itself and senses the reflected wave. The ultrasonic wave is transmitted, and the reflected wave from the vehicle and the reflected wave from the road return. The vehicle is detected by detecting the time difference until it comes. An example of such an ultrasonic detector is disclosed in Patent Document 1.

JP 60-78373 A (refer to the claims)

  In order to detect a more accurate traffic jam situation, it is desirable to obtain the vehicle speed more accurately.

However, with conventional vehicle detectors, it may be difficult to accurately determine the vehicle speed.
The ultrasonic detector detects the vehicle by receiving the reflected wave of the ultrasonic wave transmitted as described above. Therefore, when it is transmitted to a material that is difficult to reflect ultrasonic waves, such as a truck hood, it is difficult to accurately receive the reflected wave, and it is difficult to detect the vehicle. Cannot be obtained accurately.

  In addition, the vehicle detector needs to continue to detect the vehicle during a traffic jam. As shown in FIG. 22 (A), the vehicle 200 can continue to be detected if the vehicle 200 is stopped in the monitoring range formed based on the detector under the vehicle detector at the time of traffic jam. . However, as shown in FIG.22 (B), the vehicle detector is positioned between the vehicles 200, that is, when the vehicle 200 is stopped outside the monitoring range formed based on the detector, The occupation rate is detected low despite the traffic jam. The vehicle detector normally outputs the ratio of time during which the vehicle 200 occupies the road 100 as an occupancy ratio, but the occupancy ratio is not properly detected in the state shown in FIG. 22 (B). In this state, the speed of the vehicle cannot be determined appropriately.

  Furthermore, since the ultrasonic detector is an active sensor, power consumption tends to be large. For this reason, the ultrasonic detector is usually supplied with power by a cable or the like, and not only does the cable connection work for power supply be necessary, but also the power consumption is high, so it tends to be expensive. .

  Therefore, a main object of the present invention is to provide a vehicle detector that can appropriately determine the speed of the vehicle. Another object of the present invention is to provide a vehicle detection system capable of accurately detecting a vehicle.

  In the vehicle detector of the present invention, at least two thermopile elements that passively sense infrared rays are arranged so that the monitoring range based on each element is aligned in the lane direction of one lane, and the input obtained from both elements The above main object is achieved by adopting a configuration in which the speed is calculated using the level value.

  Specifically, the vehicle detector of the present invention detects the speed of a vehicle existing in a monitoring range on a road, and includes a first thermopile element and a second thermopile element that sense infrared rays emitted from a detection target, and these And a speed detection unit that calculates the speed of the vehicle using the input level values obtained from both elements. Both thermopile elements are arranged such that the monitoring range based on each element is aligned in the lane direction of one lane. The speed detection unit calculates a time difference of the output waveform of the input level value obtained from each element, a speed calculation that calculates the speed of the vehicle from the obtained time difference and the distance between the two monitoring ranges. Means.

  Conventionally, in an ultrasonic detector that is often used for detecting a vehicle, it is difficult to accurately receive a reflected wave depending on a material that receives the ultrasonic wave, and it may be difficult to detect the vehicle and the speed.

  In addition, the ultrasonic detector is generally installed so that the transmitter / receiver is located almost directly above the vehicle passing through the road so that the wave can be transmitted substantially perpendicularly to the vehicle passing through the road. There is a need to. Specifically, a horizontal member is attached to a pillar near the road, and is fixed to the horizontal member so that the transmission / reception unit of the ultrasonic detector is almost directly below the road or the vehicle. For this reason, the horizontal member needs to be long enough to protrude to the middle of the road so that the ultrasonic detector is located almost directly above the vehicle. Therefore, when using an ultrasonic detector, a relatively long horizontal material is required for installation, and this horizontal material protrudes upward in the middle of the road and is attached to a support column, so that the appearance is not impaired. There is no installation cost.

  Therefore, in order to improve the aesthetics and reduce the installation cost, the horizontal member is removed or shortened, and it is attached to the support so that the vehicle detector is located obliquely above the vehicle, instead of the so-called direct installation. It can be considered to install in a so-called side fire type. However, when installed in the sidefire type, the ultrasonic detector is directed obliquely downward with respect to the road and vehicle, so that the ultrasonic detector is affected by wind and rain and the influence of multiple reflections. There is a risk that the vehicle speed may not be required or the vehicle may be misidentified. Specifically, when the raining rain hits the ultrasonic vibrator and the vibrator vibrates at the natural vibration frequency, or received a reflected wave returned by a multipath other than the reflected wave returned directly from the vehicle. In some cases, it may be determined that there is a vehicle. Therefore, if the ultrasonic detector is installed in a so-called sidefire type in order to improve the aesthetic appearance, it may be difficult to detect the vehicle and the speed with high accuracy, or the vehicle may not be detected.

  On the other hand, the amount of infrared rays emitted from an object is proportional to the fourth power of the absolute temperature of the object and proportional to the emissivity ε of the object according to Stefan-Boltzmann's law. The emissivity ε of an object on the road, for example, a road surface or a vehicle traveling on the road, is usually substantially the same (usually 0.9 or more), and there is often no significant difference. Therefore, if a sensor that detects infrared rays is installed in the direction of the road surface, when a vehicle having a temperature different from that of an object other than a vehicle such as a road passes through the road surface, the amount of infrared light that the sensor detects is reduced. By changing, the vehicle can be detected. Further, the speed of the vehicle can be obtained using data obtained from this sensor. Therefore, the detection target is emitted rather than the detection using ultrasonic waves so that the vehicle and the speed can be detected regardless of the material of the vehicle, and the vehicle and the speed can be accurately detected even if the vehicle is installed in a so-called sidefire type. It is conceivable to detect a vehicle based on a temperature difference between the road surface and the vehicle, that is, a difference in the amount of infrared rays, using a sensor that detects infrared rays. As such a sensor, in the present invention, a sensor that does not sense the infrared ray emitted by itself but senses the infrared ray emitted from the detection target, a so-called passive sensor is used. Specifically, there is a sensor that can detect a change in electrical properties caused by a rise in temperature that is warmed by the thermal effect of infrared, and in particular, a thermopile element that outputs a temperature change generated in a thermocouple by infrared as a thermoelectromotive force. Can be mentioned.

  On the other hand, when a certain road is considered as a section over a long distance, there are usually a plurality of areas where vehicle detection is desired on the road. Conventionally, since one vehicle detector is provided for each lane in one area, one monitoring range is formed in one lane in the one area. Therefore, conventionally, a vehicle passing through one lane in one area is detected by one vehicle detector, that is, in one monitoring range. However, as a result of various studies by the present inventors, one vehicle detector having one thermopile element is arranged per lane as in the conventional case, and one monitoring range is formed based on this element. When the vehicle was detected, it was found that sometimes it was difficult to detect the vehicle.

  In each part of a vehicle such as an engine unit, a tire, a roof, etc., there is a part where a temperature difference from a road surface is generated mainly by a difference in heat capacity, a difference in thermal conductivity, or a difference in heat absorption rate of sunlight. For example, a roof is applicable. In such areas, it is known that there are two times a day when the temperature difference from the road surface almost disappears (Infrared Engineering, Editor: IEICE, Author: Haruyoshi Kuno, 1994) (See the first edition of March 20, pp. 136, Figure 6.3). Therefore, when a specific part of the vehicle passing through the monitoring range is monitored by one monitoring range and infrared rays from this part are detected, it is difficult to detect the vehicle in the above time zone depending on the part. There is. In particular, in the case of rain or snow, a water droplet absorbs a difference in thermal conductivity or a difference in thermal emissivity, and the difference in thermal conductivity is easily reduced. Therefore, in the time zone when there is almost no temperature difference from the above road surface, the difference in thermal conductivity and thermal emissivity is small from the beginning, and water drops absorb these differences, and more and more The temperature difference is likely to be small. When only the same part is monitored in the vehicle as described above, depending on the part, it is difficult to accurately detect the vehicle and it is difficult to calculate the speed of the vehicle. In particular, depending on the weather such as rain or snow, it becomes more difficult to detect the vehicle. In view of these circumstances, it is desired to monitor a portion of a vehicle where a temperature difference from the road surface is likely to occur even in weather such as rain or snow regardless of the time zone.

  Therefore, as shown in FIG. 23, it is conceivable to expand the monitoring range 401 of the vehicle detector 400 attached to the column 300 near the road 100 so as to cover the entire lane. At this time, the thermopile element can detect infrared rays from a plurality of different parts of the vehicle 200. However, even if the size of the monitoring range 401 is expanded as shown in FIG. 23 and infrared rays are detected from the entire lane, the output voltage of the thermopile element is the fourth average of the absolute temperature of the entire monitoring range. For this reason, the S / N ratio (signal to noise ratio) is lowered. Therefore, there are cases where the output result is the same as that on the road surface (time zone), which is the same as the case where infrared rays from one part of the vehicle are sensed.

  Here, each part of one vehicle, that is, an engine part, a tire, a bonnet, a side surface, a roof, and the like are usually different from each other in a uniform temperature. Moreover, the main cause which produces the temperature difference with a road surface may differ with each site | part. For example, roofs and side surfaces are sites where temperature differences from the road surface occur mainly due to differences in heat capacity, thermal conductivity, and heat absorption rate of sunlight. This is a part that tends to cause a temperature difference from the road surface.

  Therefore, it is difficult to detect a vehicle by using a plurality of elements to provide a plurality of monitoring ranges in one lane, rather than using one thermopile element to expand the size of the monitoring range in one lane. Reduce the number of cases. That is, in a configuration including a plurality of thermopile elements so as to form a plurality of monitoring ranges for one lane in an area where vehicle detection is desired provided on the road, a plurality of vehicles in one vehicle passing through these monitoring ranges are provided. Infrared rays emitted from different parts can be detected by different elements.

  Therefore, the configuration in which a plurality of monitoring ranges in the above one lane are provided is such that even if a temperature difference (difference in the amount of infrared rays) between a certain part and the road surface is less likely to occur in one vehicle (time zone), other parts It is possible to increase the possibility that a temperature difference is obtained. In addition, in the configuration in which a plurality of monitoring ranges are provided in one lane, a part that stably shows high temperature even in weather such as rain or snow, that is, a tire, an engine unit, a muffler, or the like that has a strong tendency to increase the amount of infrared rays. The possibility that the part can be detected is increased. Furthermore, by providing a plurality of thermopile elements, the monitoring range in one lane per element is relatively small, and a plurality of such monitoring ranges are provided, so that the monitoring range in one lane can be increased with one element. In this case, the detection level is not lowered due to the temperature averaging. Therefore, in this configuration, it is possible to effectively reduce cases where it is difficult to detect the vehicle, and it is possible to detect the vehicle and detect the speed with high accuracy without lowering the detection level.

  In addition, in a configuration using a passive sensor with low power consumption such as a thermopile element, power consumption can be reduced and cost can be reduced. Conventional ultrasonic detectors generally consume a large amount of power, so that it is usually difficult to supply power using a solar cell or the like, and power supply by wire is necessary. On the other hand, when a passive sensor is used, power consumption is small even when the sensor is continuously operated, and power can be supplied by a power source such as a solar cell. In addition, since this configuration does not require power supply by wire, no cable connection work is required.

  Based on the above knowledge, the vehicle detector of the present invention includes a plurality of thermopile elements so as to form a plurality of monitoring ranges in one lane. In particular, the monitoring ranges based on each element are arranged side by side in the lane direction of one lane. By providing two elements as described above, infrared rays from the vehicle can be detected more reliably. By providing such a thermopile element and a speed detector, the vehicle detector of the present invention can accurately determine the speed of the vehicle. The present invention will be described in detail below.

  When multiple thermopile elements are provided, the form of arranging a plurality of monitoring ranges on the road surface of one lane includes the form of arranging in the lane direction of one lane, the form of arranging in the lane direction of one lane, the width of one lane The form which shifts and arranges in a direction and a lane direction is mentioned. In the present invention, it is arranged at least in the lane direction. In either the vehicle width direction or the lane direction, each monitoring range may be provided so as to partially overlap with other monitoring ranges, may be provided adjacent to each other, or may be provided with an appropriate interval. May be provided. Further, a plurality of monitoring ranges may be provided in any of the vehicle width direction and the lane direction. In the present invention, at least two monitoring ranges are provided in the lane direction. Each monitoring range is provided in one lane so that at least different parts of one vehicle can pass through. Further, each thermopile element is provided in the vehicle detector of the present invention so that each monitoring range is located as described above.

  The speed detection unit includes time difference calculation means and speed calculation means. The time difference between the two output waveforms calculated by the time difference calculating means may be obtained by obtaining a correlation function from both output waveforms and obtaining the correlation function based on the correlation function. More specifically, for example, the maximum value of the correlation function is used. The distance between the two monitoring ranges calculated by the speed calculation means is obtained using, for example, the arrangement position of the vehicle detector and the difference in the directivity angles of both thermopile elements. The obtained distance is stored as a set value in the memory of the speed detector so that the speed calculation means can be called.

  In addition, the detector of the present invention can be provided with an infrared transmission lens in front of the detection direction of the element so that the infrared detection range can be narrowed and adjusted to an appropriate range, and the infrared ray can be easily collected on the thermopile element. preferable. The infrared transmitting lens is not particularly limited as long as it transmits infrared light. For example, one surface may be spherical. The infrared transmitting lens is particularly preferably made of ZnS. Conventional infrared transmission lenses made of Ge (germanium) or the like are known, but conventional lenses require glass-based or silicon-based auxiliary materials. However, as a result of studies by the present inventors, a lens made of ZnS has excellent weather resistance, and it has been found that even if it is exposed to the outside, it can withstand sufficient use. Such a lens may be provided one by one according to the number of thermopile elements, or a plurality of elements may be arranged for one lens. In the former case, the arrangement state of the monitoring range formed by each element can be changed by changing the directivity angle of each lens while the thermopile element is arranged along the central axis of the lens. Also, the arrangement state of the monitoring range formed by each element can be changed by changing the arrangement position of the thermopile element with respect to the central axis of the lens. In the latter case where a plurality of elements are arranged in one infrared transmission lens, the arrangement state of the monitoring range formed by each element can be changed by shifting the position of each thermopile element with respect to the central axis of the lens. In this case, the number of infrared transmission lenses can be reduced.

  In addition, the detector of the present invention preferably includes a housing for storing the thermopile element. Such a casing is preferably formed of lightweight aluminum or the like. The infrared transmitting lens is disposed in the casing so that the focal length matches the thermopile element housed in the casing. At this time, a support part that supports the thermopile element and a support part that supports the infrared transmission lens may be provided separately in the housing, but may be an integrally formed support part. In the case of an integrally formed support portion, the element and the lens arrangement location may be formed on the support portion so that when the thermopile element and the infrared transmission lens are arranged, the focal length becomes appropriate. Is preferred. When such a support portion is used, the thermopile element and the infrared transmissive lens are appropriately disposed at each arrangement position, thereby obtaining an integral member that can appropriately adjust the focal length, and arranging these at predetermined positions on the housing. At this time, it is not necessary to adjust the focal length, and it is preferable because it can be easily installed in the housing.

  Moreover, it is preferable that the housing includes an aiming unit for adjusting the directivity angle of the infrared transmitting lens in a target direction. When a case having a thermopile element and an infrared transmission lens is installed on a pillar near a road, it is necessary to adjust the directivity angle of the lens to a desired direction. Therefore, when a casing having an aiming unit is used, the directivity angle of the infrared transmitting lens arranged in the casing can be easily grasped, and the installation workability is good. The aiming unit may be any mark that can adjust the directivity angle in a desired direction. For example, a mark such as a combination of a concave protrusion and a convex protrusion, or a laser pointer or the like may be provided. A configuration is mentioned. In the former case, more specifically, one surface of the housing has a concave projection at one end and a convex projection at the other end, and confirms the convex projection from the recess of the concave projection, and the convex portion of the concave projection and the convex projection. The structure which can make a directivity angle into an appropriate direction by matching the straight line which connects with the target direction is mentioned.

  The casing may be installed in a so-called side fire type with respect to a support provided on the road, and infrared detection may be performed from the side of the road. Specifically, when the mounting position of the housing is higher than the height of the vehicle from the road surface, it is mounted on the column so that the housing (detection direction of the thermopile element) faces diagonally downward with respect to the road or vehicle. May be detected obliquely from above. By using a thermopile element that senses infrared rays, the detector of the present invention can accurately sense infrared rays from the vehicle even when installed on the side of the road, not just above the vehicle passing through the road. The vehicle speed can be obtained with high accuracy. Further, the detector of the present invention does not need to use a horizontal member at the time of installation like a conventional ultrasonic detector, or even if a horizontal member is used, it may be shorter than a conventional horizontal member. , Less aesthetic loss.

  Construction of a system comprising the vehicle detector of the present invention and a vehicle presence / absence determination unit for determining the presence / absence of a vehicle present in the monitoring range using the input level value obtained from each thermopile element included in the detector. By doing so, in addition to detecting the speed of the vehicle, there is little misidentification of the vehicle, and the vehicle can be detected with high accuracy. Since this system is provided with a plurality of thermopile elements and has a plurality of monitoring ranges for one lane, it is possible to detect different parts of one vehicle, particularly parts having different temperatures, in each monitoring range. . Therefore, this system can detect a vehicle with high accuracy even in an environment such as temperature or weather or in a traffic jam. In particular, by constructing a vehicle detection system of the present invention that includes an identity determination unit that determines the identity of a vehicle using the vehicle speed calculated by the detector of the present invention, which will be described later, the misidentification of the number of vehicles is reduced. it can.

  In a conventional vehicle detection system using a vehicle detector, the presence or absence of a vehicle is determined based only on data obtained from one sensor. On the other hand, in the system of the present invention, a plurality of monitoring ranges are provided in one lane in one area where vehicle detection is desired, input level values are obtained from a plurality of thermopile elements, and a plurality of obtained data (input The presence or absence of the vehicle is determined using the level value. Therefore, the detection accuracy of the vehicle can be increased.

  In particular, the system of the present invention including the detector of the present invention that arranges a plurality of monitoring ranges in the lane direction of one lane may cause the vehicle to stop in any of the monitoring ranges aligned in the lane direction even in a traffic jam. Becomes higher. Therefore, the possibility that any one of the thermopile elements forming the monitoring range detects the vehicle can be increased. Therefore, it is possible to further improve the accuracy of counting the number of vehicles in a traffic jam and the accuracy of measuring the occupation rate. The occupancy rate is the ratio of vehicles occupying the road. When this value is large, it indicates that the vehicle is congested.

  Further, the system of the present invention in which a plurality of monitoring ranges are arranged in the lane direction of one lane can reduce misidentification of the number of vehicles. In one vehicle, for example, there are relatively many vehicles in which the temperature at the top and the rear is high and the temperature difference between the middle and the road surface is small. At this time, two waveforms appear continuously in the data obtained from each thermopile element. When such a vehicle travels at a low speed, there is a risk of detecting one vehicle as two from two waveforms. However, the system of the present invention in which a plurality of monitoring ranges are provided in the lane direction can determine whether the two waveforms are of the same vehicle by the identity determination unit described later, and can accurately distinguish them.

  As the thermopile element used in the present invention, in particular, a thermopile element having a large thermoelectromotive force output is used, and another sensor such as a pyroelectric sensor is used by devising an algorithm for determining the presence or absence of a vehicle as described later. Without using, it is possible to detect the vehicle by performing sufficient infrared sensing only with the thermopile element.

  Furthermore, if the circuit configuration is devised as will be described later, the vehicle can be detected with high accuracy even when the vehicle is stopped due to a traffic jam or a stagnation. A pyroelectric sensor, on the other hand, generally produces a large output, but is an element that outputs a voltage corresponding to a change in the temperature of a measurement target. It is difficult to perform proper detection.

  Examples of the vehicle presence / absence determining unit include a configuration including the following means. That is, in the input level value obtained from each thermopile element, the value based on the amount of infrared rays emitted by an object other than the vehicle is set as the background level of each monitoring range, and the background level of each monitoring range is determined from the value based on the amount of infrared rays. A background level calculation means for calculating, and a comprehensive determination means for comparing the comparison value with a threshold value to determine the presence or absence of a vehicle by using a value based on the difference between the input level value of each monitoring range and the background level as a comparison value. Can be listed.

  Although it is conceivable to use the input level value obtained from each thermopile element as it is, in the present invention, an appropriate calculation is performed using a plurality of obtained input level values, and the presence / absence of a vehicle is determined using the calculated value. Make a decision. That is, the procedure of vehicle detection in the system of the present invention will be described. First, infrared rays are sensed by each thermopile element, and input level values are respectively obtained. Using the obtained input level value, a background level based on each input level value is calculated as described later. Further, the difference between the input level value from each thermopile element and the background level is obtained for each monitoring range. A value based on the difference between the obtained input level value and the background level is calculated and used as a comparison value. For example, when the comparison value is equal to or greater than the threshold value, it is determined that there is a vehicle, and when the comparison value is less than the threshold value, it is determined that there is no vehicle. Thereafter, the obtained determination result is transmitted to a signal controller or a management center.

  The background level is a value based on the amount of infrared rays emitted by an object other than the vehicle, and is calculated using the input level value as described above. In the present invention, since the measurement data is used as the background level, the background level can be approximated to the actual environment value, and more accurate detection can be performed.

As an operation value using the input level value, an exponential smoothing method using a past background level can be cited. Exponential smoothing is generally _{f 0 = α × d -1 +} (1-α) is expressed as _{f -1 = f -1 + α ×} (d -1 -f -1) (f 0: next predicted value, alpha : Smoothing coefficient, d ₋₁ : actual value of the previous period, f ₋₁ : predicted value of the previous period), and actual value of the previous period (here, input level value) can be reflected. Therefore, the background level can be a more accurate value in accordance with the actual environment (road surface condition). More specifically, the background level is made to follow the input level value at a follow-up speed determined by the smoothing coefficient α. The smoothing coefficient α may be a constant value, but it is preferable to change it according to the previous vehicle determination result because the background level can be more reliably grasped regardless of the amount (temperature) of the vehicle infrared rays. . For example, when the previous vehicle determination result is that there is a vehicle, the current input level value is considered to be a value obtained from infrared rays from the vehicle when there is a traffic jam. Accordingly, when the follow-up speed is increased in a traffic jam or the like, an abnormal background level is obtained. Therefore, it is preferable to make the follow-up speed relatively low or zero. That is, the smoothing coefficient α is made relatively small or 0. At this time, the previous background level is used as it is with little involvement of the actual value of the previous period (here, the input level value). On the other hand, when the previous vehicle determination result is that there is no vehicle, the current input level value is considered to be a value obtained from infrared rays from other than the vehicle, that is, from the road. Therefore, it is preferable to make the correction by the actual value of the previous period (here, the current input level value) by relatively increasing the follow-up speed, that is, by relatively increasing the smoothing coefficient α.

  As the comparison value, for example, the total sum of the difference between the input level value for each monitoring range and the background level (hereinafter referred to as background difference) may be used. By taking the sum, it is possible to detect the vehicle if there is a temperature change in at least one monitoring range, and increase the change even if there is only a minute temperature change in all the monitoring ranges. Therefore, the vehicle can be detected more reliably. The sum of background differences of each monitoring range may be used as it is.However, as a result of investigations by the present inventors, when the amount of infrared rays is gently changed during a traffic jam or when the temperature of the vehicle (infrared rays is changed). In some cases, such as when the input level value is similar to the background level, there is a possibility that it may be determined that there is no vehicle. In such a case, it is effective to use a value obtained by integrating the input level value within a certain time, not an instantaneous input level value. It was also possible to grasp the essential trends in the change in quantity and to obtain more accurate vehicle detection. Therefore, it is proposed to use a value obtained by integrating the total sum of background differences in each monitoring range for a certain time as a comparison value.

  In addition, as a comparison value, in addition to the sum of the background differences, it was also found that it is preferable to use the amount of change per unit time of each input level value. Therefore, it is proposed to use the change amount as a comparison value. In particular, it is preferable to use the total amount of change in each monitoring range as a comparison value because the vehicle can be detected more reliably as described above. The amount of change may be, for example, the difference between the previous input level value and the current input level value in each monitoring range, but the difference between the previous input level value and the current input level value, for example, every 10 ms. When measuring the input level value, it is more effective to make the difference from the input level value before 160 ms. Since this amount of change is not calculated with the background level taken into account, it is not affected by the background level. Therefore, even if the background difference is small, the vehicle exists while the input level value is changing. It is easy to obtain a determination. Therefore, the case where the vehicle cannot be recognized as described above is reduced. The determination of the presence / absence of the vehicle based on the comparison value using the change amount is performed, for example, by performing the first determination based on the sum of the background differences of the respective monitoring ranges, and further performing the second determination based on the sum of the change amounts. You may go by. Specifically, if the sum of the background differences is equal to or greater than the threshold, it is determined that there is a vehicle. In addition, the presence of the vehicle can be detected more accurately.

  Furthermore, as the comparison value, it is more preferable to use a value obtained by integrating the sum of the background differences for a certain period of time and the amount of change. If the integrated value is used alone as a comparison value, this value may become 0 (zero) .However, while the input level value is still changing, there is a vehicle by using the change amount. It is easy to obtain a determination that In this way, using a thermopile element with a large output and using not only the integrated value but also a value that also considers the amount of change in the vehicle determination algorithm, only a thermopile element is provided without a separate sensor such as a pyroelectric sensor. However, it can have a function equivalent to or better than that of a pyroelectric sensor.

  The threshold (threshold) used for the determination of the presence or absence of the vehicle may be a constant value, but is preferably changed according to the actual environment. For example, when the change in the amount of infrared light (temperature change) is large, that is, when the dispersion is large, the threshold value is set to a relatively large value, and when the change in the amount of infrared light (temperature change) is small, that is, when the dispersion is small. The threshold may be a relatively small value. These threshold values may be obtained by calculation. For example, the threshold value may be changed by setting the threshold value as a set value + correction value and changing the correction value. The correction value is preferably changed based on the input level value. For example, the sum of the background differences may be used. Further, it is preferable that the correction value is changed according to the previous vehicle determination result because the threshold value can be prevented from becoming too large. For example, when the previous vehicle determination result is that there is a vehicle, it is preferable that the correction value is relatively small, and when the previous vehicle determination result is that there is no vehicle, the correction value is relatively large. In this case, once it is determined that the vehicle is present, it is easy to continue the determination that the vehicle is present. For example, when the vehicle is stopped due to traffic jam or the like, the determination that the vehicle is present is continued and the vehicle starts running. At this time, the determination that the vehicle is present can be easily stopped.

As another vehicle presence / absence determination unit, for example, a configuration including the following means can be given.
I. In the input level value obtained from each thermopile element, the value based on the amount of infrared rays emitted by an object other than the vehicle is set as the background level of each monitoring range, and the background level of each monitoring range is determined from the value based on this amount of infrared rays. Background level calculation means for calculating.
II. The sum of the input level values obtained from each thermopile element is taken, the sum of the background levels obtained from the background level calculating means is taken, and the sum of the input level values and the sum of the background levels are calculated. First determination means for determining the presence of the vehicle by using a value based on the difference as a first comparison value and comparing the first comparison value with a first threshold value.
III. The difference between the input level values obtained from each thermopile element is taken, the difference between the background levels obtained from the background level calculation means is taken, and the difference between the input level value and the difference value between the background levels is calculated. Second determination means for determining the presence of the vehicle by using a value based on the difference as a second comparison value and comparing the second comparison value with a second threshold value.
IV. Comprehensive determination means for determining the presence or absence of a vehicle based on the sum of both the first presence result of the first determination means and the second presence result of the second determination means.

  In this configuration, the sum and difference of the input level values obtained from each of the plurality of thermopile elements are taken, and based on these sum and difference values, the vehicle presence status is determined by the first determination means and the second determination means, respectively. judge. Then, based on the result based on the sum of the input level values (first presence result) and the result based on the difference between the input level values (second presence result), the comprehensive determination means finally determines the presence or absence of the vehicle. That is, the first and second determination means determine the presence of the vehicle by directly using the data obtained from each thermopile element, and this result is used as a temporary result, and the comprehensive determination means uses this temporary result. Thus, the determination is made in multiple stages, such as determining the presence or absence of a vehicle. This configuration effectively reduces cases where vehicle detection is difficult.

  Further, in the configuration using the sum of the input level values, the noise can be reduced by about 1.4 times and the signal can be increased by about 2, so that the S / N ratio can be increased to reduce the noise. Therefore, this configuration can more accurately determine the first presence result. Further, in this configuration, the difference in the input level value from each thermopile element can be confirmed by using the difference value in the input level value. Depending on the part of the vehicle, the input level value may be smaller than the background level, and the sum of the input level value larger than the background level and another input level value smaller than the background level may be almost zero. Conceivable. Therefore, it may be difficult to determine the presence of the vehicle only with the sum of the input level values. Therefore, in the present invention, by using not only the sum of the input level values but also the difference value of the input level values, the input level value becomes the same as the background level (the amount of infrared rays from the road surface) and the vehicle The case where it is difficult to detect can be reduced effectively.

  The vehicle detection procedure in the system of the present invention having the above configuration will be described. First, infrared rays are sensed by each thermopile element, and input level values are obtained to obtain the sum and difference. On the other hand, based on each input level value, the background level is calculated in the same manner as in the above algorithm, and the sum and difference of the background levels are taken. When a value based on the difference between the sum of the obtained input level values and the sum of the background levels is calculated to be a first comparison value, and the first comparison value is equal to or greater than the first threshold value in the first determination means It is determined that there is a vehicle, and if it is less than the first threshold, it is determined that there is no vehicle. Similarly, when a value based on the difference between the difference value of the input level value and the difference value of the background level is calculated to be a second comparison value, and the second comparison value is equal to or greater than the second threshold value in the second determination means It is determined that there is a vehicle, and if it is less than the second threshold, it is determined that there is no vehicle. Then, the sum of the results from the first determination means and the second determination means is calculated, and the comprehensive determination means finally determines the presence or absence of the vehicle. Thereafter, the obtained determination result is transmitted to a signal controller or a management center.

  The sum of input level values may be the sum of all input level values, ie, the sum. The difference between the input level values is the value obtained by subtracting all the other input level values from one input level value, that is, if there are three input level values, A, B, and C, A-B-C Good. The same applies to the sum value and difference value of the background level described later.

  As in the above algorithm, a calculated value by exponential smoothing using a past background level may be used so that the background level becomes a more accurate value in accordance with the actual environment (road surface condition).

  Also, the difference between the difference between the input level value as the first comparison value and the difference between the background level and the value based on the difference between the sum value of the input level value and the background level as the second comparison value. As in the above algorithm, it is preferable to use a value obtained by integrating the sum values (difference values) of input level values within a fixed time, rather than using the difference as it is, as in the above algorithm. At this time, the vehicle can be detected more accurately. Further, as the first comparison value and the second comparison value, the change amount per unit time in the sum value of the input level values and the change amount per unit time in the difference value of the input level values may be used as in the above algorithm. preferable. More preferably, the first comparison value and the second comparison value are a combination of the integrated value and the change amount. In this way, using a thermopile element with a large output, and using a value that considers not only the integrated value but also the amount of change as a comparison value, even with a thermopile element alone as in the above algorithm, it is equivalent to or better than a pyroelectric sensor. Can have functions.

  The first threshold value and the second threshold value are preferably changed corresponding to the actual environment in the same manner as the above algorithm. For example, the correction value may be changed by setting value + correction value. The specific method of changing is preferably the same as the above algorithm.

  The comprehensive determination means comprehensively determines the presence or absence of the vehicle based on the existence results from the first determination means and the second determination means. Two existence results, a first existence result based on the sum value of the input level values and a second existence result based on the difference value of the input level values, are sent to the comprehensive determination means. In the present invention, the presence or absence of the vehicle is determined by taking the sum (or) of these existence results. That is, not only when both the first existence result and the second existence result are with a vehicle, but when one of the first existence result and the second existence result is with a vehicle and the other is with no vehicle, The comprehensive determination unit determines that there is a vehicle, and determines that there is no vehicle when both the first and second presence results are that there is no vehicle.

  Furthermore, in the system according to the present invention, in the above two algorithms, in addition to performing the vehicle presence / absence determination based on the input level value from each thermopile element, whether the obtained input level value is for the same vehicle. An identity determination unit for determining the above is provided. Specifically, the speed of the vehicle and the distance between the vehicles are calculated based on a plurality of input level values, and if the distance between the vehicles at the speed is less than or equal to a limit value, it is determined that they are the same vehicle. In particular, the vehicle speed is determined by the vehicle detector of the present invention. More specifically, the identity determination unit uses the time difference between the adjacent first waveform and the second waveform in any of the output waveforms of the input level values obtained from both elements, and the detector of the present invention. The inter-vehicle distance calculation means for calculating the distance between the first vehicle corresponding to the first waveform and the second vehicle corresponding to the second waveform from the calculated speed is compared with the calculated inter-vehicle distance and the set value. And identity determining means for determining whether or not the first vehicle and the second vehicle are the same. By such a determination, the system of the present invention has a plurality of thermopile elements and provides a plurality of monitoring ranges per lane in the lane direction, and it is difficult to detect a vehicle by performing an operation based on the above algorithm. In addition to being able to reduce this, it is possible to reduce misidentification of one vehicle as two or more vehicles.

  Further, the system of the present invention, in the above two algorithms, in addition to performing the presence / absence determination of the vehicle based on the input level value from each thermopile element, when the vehicle presence / absence determination unit determines that the vehicle is present, The speed detector of the present invention can be configured to calculate the speed of the vehicle.

  In the system of the present invention, when shifting from the presence of the vehicle to the state without the vehicle, it is desirable to extend the determination that the vehicle is present by giving a certain holding time. Depending on the vehicle, the temperature distribution varies, and the amount of infrared rays also varies due to the temperature variation, so there is a risk of misidentifying one vehicle as two or more. For this reason, misconfiguration can be effectively suppressed by adding a certain holding time. Such a compensation can be realized because it is practically impossible that the vehicle is eliminated only in a very short time, for example, it is impossible for a plurality of vehicles to travel at a high speed with a relatively short inter-vehicle distance. An example of the holding time is about 115 ms.

  The above-described background level calculation, comparison value calculation, threshold value calculation, and other vehicle presence / absence determination processing may be performed using a known central processing unit (CPU) or the like. One CPU may be provided for each thermopile element, or one CPU may be provided for a plurality of thermopile elements. In addition, the vehicle presence / absence determination unit that performs these calculations and determinations may be housed together with the thermopile element in a housing included in the detector of the present invention, or may be housed in a separate control box or the like and attached with the detector of the present invention. You may install in the vicinity of this support column or this support column.

  In the system of the present invention, the basic circuit includes a plurality of thermopile elements, an amplifier that amplifies the electromotive force from these elements, and a determination means that reads the amplified voltage and determines the presence or absence of the vehicle using this value (for example, , CPU, etc.). In such a circuit, the amplifier is configured to amplify not only the AC component but also the DC component, so that it is difficult to detect the vehicle when the vehicle is stopped due to a traffic jam or the like. In contrast, it is possible to continue to detect a stopped vehicle. In particular, a plurality of amplifiers are provided for one thermopile element. In each element, an amplifier A connected to the element side has a relatively large amplification factor, and an amplifier B connected to the judging means In addition, it is preferable to use a configuration with a relatively small amplification factor, and the amplifier B amplifies the difference between the output of the amplifier A and the reference voltage. Thus, by using a plurality of amplifiers for one thermopile element and inputting the reference voltage to the amplifier B, the dynamic range of the determination means can be improved.

  The results obtained from the comprehensive judgment means in the system of the present invention are tabulated and transmitted to a signal controller, a management center or the like. The result may be transmitted by wire, but the system may further include a wireless communication unit that wirelessly transmits the aggregation result, and may be transmitted wirelessly. At this time, it is preferable to provide a power supply control unit that intermittently supplies power to the wireless communication means, and to supply power only for a fixed time, because the power can be further reduced. For example, the power control unit is provided with a timer or the like so that the power is turned on only for a certain time in a certain cycle, and the counting result is received and transmitted only when the power is turned on.

  As described above, according to the vehicle detector of the present invention, the speed of the vehicle can be obtained more accurately, which can contribute to more accurate detection of a traffic jam situation. The system of the present invention including such a detector of the present invention can effectively reduce the case where it is difficult to detect the vehicle and the misidentification of the number of vehicles.

Embodiments of the present invention will be described below.
<Vehicle detector>
The vehicle detector of the present invention includes two thermopile elements that sense infrared rays emitted from detection targets (vehicles, roads, etc.), and a speed detector that calculates the speed of the vehicle using input level values obtained from both elements. Prepare. In particular, the vehicle detector includes both elements so that a monitoring range based on each thermopile element is arranged in the lane direction with respect to one lane. The speed detection unit is a time difference calculating means for calculating a time difference between output waveforms of input level values obtained from each thermopile element, a speed for calculating a vehicle speed from the obtained time difference and a distance between two monitoring ranges. Computation means.

  FIG. 1 is a model diagram showing an output waveform of an input level value obtained from a thermopile element. In the output waveforms 3A ′ and 3B ′ of the input level value from the thermopile element, the waveforms A and B are due to the vehicle. The said vehicle detector calculates the speed of a vehicle using this waveform. Specifically, first, a distance x between two monitoring ranges provided in the lane direction is obtained in advance, and this distance x is stored as a set value in the memory of the speed detection unit. Then, the time difference calculating means calculates the time difference t between the output waveforms 3A ′ and 3B ′ of the input level value based on each thermopile element. More precisely, a correlation function is calculated from waveform A (waveform B) of both output waveforms 3A ′ and 3B ′, and the maximum value of the correlation function is obtained. Next, the speed calculation means calculates the vehicle speed v = x / t from the distance x called from the memory and the time difference t obtained by the calculation.

  The vehicle detector of the present invention can determine the speed of the vehicle more accurately. Therefore, when this speed is used, it is considered that a more accurate traffic jam situation can be detected. Further, by providing this detector in a vehicle detection system described later, it is possible to detect the vehicle and to determine not only the speed of the vehicle but also the identity. When constructing a vehicle detection system, which will be described later, it is understood that the waveforms A and B of the output waveform are based on the vehicle when it is determined that there is a vehicle. Therefore, the vehicle detector can be configured to calculate the speed of the vehicle when it is determined that the vehicle is present.

Next, the vehicle detection system of the present invention will be described.
<Overall configuration of vehicle detection system>
FIG. 2 is a functional block diagram of the vehicle detection system of the present invention. The vehicle detection system 1 of the present invention includes a vehicle detector (not shown) having a thermopile element 3 for detecting infrared rays emitted from a detection target (vehicle, road, etc.), and a detection result (input level value) obtained from the element 3. A vehicle presence / absence determination unit 10 that performs a sensing process for determining the presence / absence of a vehicle using an output, and an identity determination unit (not shown) that determines the identity of the vehicle using an output waveform of an input level value obtained from the element 3 ). Particularly, in the present invention system, a plurality of thermopile elements 3 are provided, a plurality of monitoring ranges based on each element 3 are provided for one lane, and a plurality of elements 3 are detected for one vehicle existing in these monitoring ranges. It is.

  In the detection system of the present invention shown in FIG. 2, the wireless communication unit 11 for transmitting the result from the vehicle presence / absence determination unit 10 to a signal controller, a management center, etc., the sensor unit 2 having the thermopile element 3, and the determination unit 10 The solar cell unit 12 is provided for power supply, and power is supplied by the solar cell 12a.

  With this configuration, infrared rays from the detection target are sensed by a plurality of thermopile elements 3 provided in the sensor unit 2 in the vehicle detector, and the electromotive force generated in each element 3 is amplified by the amplifier 13 to detect the presence / absence of the vehicle 10 and converted into digital signals by the A / D converter 14 to obtain the input level value from each element 3. Then, the vehicle presence / absence determination unit 10 determines the presence of the vehicle based on whether the calculated value using the obtained plurality of input level values is equal to or greater than a threshold value. The vehicle detector calculates the speed using the obtained input level value. Further, the identity determination unit determines the identity of the vehicle using the obtained input level value. The obtained results are aggregated and sent to the wireless communication unit 11, and the aggregated results are sent to a signal controller, a traffic management center, etc. via the transmission / reception unit 11a. Although FIG. 2 shows the case where the result is transmitted to the management center or the like via the wireless communication unit, it may be performed via a wire such as a cable. The components of the system of the present invention, the algorithm for determining the presence / absence of a vehicle, and the determination procedure for identity will be described in detail below.

(Example 1: When the monitoring range is arranged in the lane direction)
《Thermopile element and monitoring range》
FIG. 3 (A) is a front view schematically showing the vehicle detection system of the present invention that forms two monitoring ranges in the lane direction, (B) is a side cross-sectional view along B-B, and FIG. 4 is this vehicle detection. It is explanatory drawing which shows the state which attached the system to the support | pillar of the roadside. Hereinafter, the same reference numerals in the drawings denote the same items. The vehicle detection system 1a of the present invention is obtained from a sensor unit 2 having a thermopile element 3a, 3b that senses infrared rays emitted from the side of the road 100 and from the side of the road 100, and the elements 3a, 3b. A vehicle presence / absence determination unit 10 having a CPU board 5c for determining the presence / absence of the vehicle using the obtained data. Further, the system 1a includes a speed detection unit that calculates the speed of the vehicle using data obtained from the elements 3a and 3b, and an identity determination unit (none of which is shown) that determines the identity of the vehicle. Prepare. This vehicle detection system 1a is attached to a column 300 near the road 100 and detects a vehicle 200 passing through monitoring ranges 3A and 3B formed based on the thermopile elements 3a and 3b of the sensor unit 2. In particular, the sensor unit 2 includes two thermopile elements 3a and 3b, and the monitoring ranges 3A and 3B formed based on the elements 3a and 3b are adjacent to each other in the lane direction in one lane of the road 100 ( Both elements 3a and 3b are arranged. The arrangement of the thermopile elements 3a and 3b will be described later.

  Such a system 1a of the present invention can increase the possibility that the vehicle stops in at least one of the monitoring range 3A or the monitoring range 3B, for example, in a traffic jam. Therefore, the system of the present invention can reduce the case where it is difficult to detect the vehicle. In particular, the system according to the present invention reduces the misperception of the number of vehicles, increases the accuracy of occupancy measurement, can more accurately determine the speed of the vehicle, and realizes a more accurate detection of a traffic jam situation.

  In addition, the vehicle detector of the present invention can be performed with high accuracy even when detecting a vehicle from the side of the vehicle by using a thermopile element instead of an ultrasonic wave in which the installation conditions are limited. Furthermore, since the thermopile element is a passive sensor with lower power consumption, it is possible to reduce power consumption even when the sensor is continuously operated.

<Case>
In the vehicle detection system 1a, the thermopile elements 3a and 3b are housed in the housing 4 as shown in FIG. 3 (B). These thermopile elements 3a and 3b are mounted on the substrates 5a and 5b, respectively, and arranged on the housing 4. Each board 5a, 5b is connected to the CPU board 5c via a connector or the like.

  The housing 4 is a box-shaped body formed from lightweight aluminum. In this example, an object having an aiming mechanism is used as will be described later. Further, a cover may be provided on the outer periphery of the housing 4.

"lens"
Infrared transmitting lenses 6a and 6b are arranged in front of the detection direction of thermopile elements 3a and 3b (left side in FIG. 3B) so that one surface (spherical surface in FIG. 3B) is exposed. In this example, a spherical lens formed of ZnS is used, and since it has excellent weather resistance, a glass-based or silicon-based auxiliary material need not be separately provided.

  The thermopile elements 3a and 3b and the infrared transmission lenses 6a and 6b are supported by a block-like support portion 7a and arranged in the housing 4. In this example, the support portion 7a supports the thermopile elements 3a and 3b and the infrared transmission lenses 6a and 6b as shown in FIG. 5, and has a through hole at a place where the elements 3a and 3b and the lenses 6a and 6b are supported. Is formed. Further, the support portion 7a is provided with an appropriate space 8 through the through-hole so as to have an appropriate focal length when the thermopile elements 3a and 3b and the infrared transmission lenses 6a and 6b are arranged, and the elements 3a and 3b and the lens are provided. It forms the installation location for 6a and 6b. Accordingly, when the thermopile elements 3a and 3b and the infrared transmission lenses 6a and 6b are arranged at predetermined installation locations of the support portion 7a, the focal length is automatically adjusted. When such a support portion 7a is used, the focal length can be easily adjusted, and the thermopile elements 3a, 3b and the infrared transmission lenses 6a, 6b are arranged on the support portion 7a and an integrated member is attached to the housing 4. At the time of mounting, adjustment of the focal length is not required, and the mounting workability to the housing 4 is excellent.

  In this example, the substrates 5a and 5b on which the thermopile elements 3a and 3b are mounted are respectively arranged on the support portion 7a and then fixed to the support portion 7a with screws 9. The infrared transmitting lenses 6a and 6b are similarly fixed to the support portion 7a with screws 9 through the lens presser 6c. In this example, two thermopile elements and two infrared transmission lenses are supported on one support part. However, two support parts are prepared, and one element and one lens are provided on each support part. You may support. The same applies to the embodiments described later.

《Arrangement of thermopile element and infrared transmission lens 1》
FIG. 6 is a schematic diagram for explaining the arrangement state of the thermopile element and the infrared transmission lens, in which each lens is provided with a lens, and (A) is an example in which the position of the element with respect to the lens is shifted, (B) is an example in which the position of the support portion is shifted. In order to effectively shift the monitoring range in one lane, it is effective to change the position of the thermopile element with respect to the infrared transmission lens or to change the position of the support part when the element and the lens are supported by an integral support part. is there. Specifically, as shown in FIG. 6 (A), when two thermopile elements 3a, 3b and infrared transmission lenses 6a, 6b are supported by one support portion 7a, the position of one element 3a is the center of lens 6a. Arranged along the axis 6d, the position of the other element 3b is shifted from the central axis 6d of the lens 6b. At this time, the direction of directivity is changed by changing the directivity angle of the thermopile elements 3a and 3b. Therefore, the positions of the monitoring ranges 3A and 3B (see FIG. 4) formed by the thermopile elements 3a and 3b can be shifted by several tens cm to several m. In this example, as shown in FIG. 3, the horizontal range is shifted, the position of the thermopile element with respect to the central axis of the infrared transmission lens is adjusted as appropriate, and the monitoring range formed by each element when the vehicle detection system is installed on the column Lined up in the lane direction. Of course, the monitoring range formed by each element can be arranged in the lane direction only by adjusting the position of the thermopile element with respect to the central axis of the infrared transmission lens.

  Further, as shown in FIG. 6B, two support portions 7a ′ and 7a ″ are prepared, the thermopile element 3a and the infrared transmission lens 6a are supported by the support portion 7a ′, and the element 3b and the lens are supported by the support portion 7a ″. 6b is supported, and when the elements 3a and 3b are arranged along the central axis 6d of the lenses 6a and 6b, respectively, the central axis 6d intersects the one supporting portion 7a ′ and the other supporting portion 7a ″. At this time, since the directivity angles of the thermopile elements 3a and 3b change in the same manner as described above, the positions of the monitoring ranges 3A and 3B (see FIG. 4) formed by the elements 3a and 3b are several tens of centimeters. It can be shifted by ~ m.

《Arrangement of thermopile element and infrared transmission lens 2》
FIG. 7 is a schematic diagram for explaining the arrangement state of the thermopile element and the infrared transmission lens, and is an example in which two elements are provided for one lens. In the vehicle detection system 1a shown in FIG. 3, an example is shown in which one infrared transmission lens is arranged for one thermopile element, but one lens is arranged for a plurality of thermopile elements as shown in FIG. May be. At this time, the monitoring range formed by each thermopile element can be shifted by changing the position of the thermopile element with respect to the infrared transmission lens. Specifically, the positions of the thermopile elements 3a and 3b are shifted from the central axis 6d of the infrared transmission lens 6, respectively. At this time, the direction of directivity is changed by changing the directivity angle of the thermopile elements 3a and 3b. Therefore, the positions of the monitoring ranges 3A and 3B (see FIG. 4) formed by the thermopile elements 3a and 3b can be shifted by several tens cm to several m. In this case, the thermopile elements 3a and 3b and the infrared transmission lens 6 are preferably supported by one support portion 7. The arrangements 1 and 2 of the thermopile element and the infrared transmission lens are the same as in Example 2 described later.

《Aiming section》
FIG. 8 is a perspective view schematically showing the vehicle detection system of the present invention. The vehicle detection system 1a of the present invention adjusts the directivity angle of the lens so that the directivity angle of the infrared transmission lens (see FIG. 3) can be easily adjusted to the target direction outside the housing 4 as shown in FIG. The sighting unit 4a is provided. In this example, on one surface of the housing 4, a concave protrusion 4b is provided at one end, and a convex protrusion 4c is provided at the other end. These protrusions 4b and 4c are provided so that the directivity angle can be adjusted appropriately when the linear direction connecting the recess of the concave protrusion 4b and the convex part of the convex protrusion 4c is the target direction. Accordingly, as shown in FIG. 4, when installing the housing 4 of the vehicle detection system 1a of the present invention on the support column 300 or the like, the operator looks into the concave protrusion 4b from the recess of the concave protrusion 4b toward the convex protrusion 4c. The installation position may be adjusted so that a straight line connecting the dent and the convex portion of the convex protrusion 4c is in the target direction. In FIG. 8, a circular hole 4d provided on the end surface of the housing 4 (left end surface in FIG. 8) is provided for projecting the infrared transmitting lens.

  In this example, an example in which the vehicle presence / absence determination unit 10 is provided in the housing 4 is shown. However, the determination unit 10 is housed in the control box, and this box is attached to the column on which the housing 4 is installed. Also good. Also, the solar cell unit 12 may be attached to a support column on which the housing 4 is installed (see FIG. 4). The same applies to the second embodiment of the vehicle detector of the present invention described later.

(Example 2: When the monitoring range is shifted in both the vehicle width direction and the lane direction)
FIG. 9 (A) is a front view schematically showing the vehicle detection system of the present invention that forms a monitoring range by shifting in the vehicle width direction and the lane direction, and FIG. 9 (B) is a side sectional view taken along line BB in FIG. These are explanatory drawings which show the state which attached this vehicle detector to the support | pillar of the roadside. The basic configuration of the vehicle detection system 1b of the present invention is the same as that of the first embodiment, and the sensor unit 2 having two thermopile elements 3a and 3b in the housing 4 and the data from the elements 3a and 3b. A vehicle presence / absence determination unit 10 having a CPU board 5c for determining the presence / absence of a vehicle Further, the system 1b includes a speed detection unit that calculates the speed of the vehicle using data obtained from the elements 3a and 3b, and an identity determination unit (none of which is shown) that determines the identity of the vehicle. Prepare. The elements 3a and 3b and the lenses 6a and 6b are supported by the support portion 7b. In the vehicle detection system 1b, in particular, both the elements 3a, 3b are arranged such that the monitoring ranges 3A, 3B formed based on the thermopile elements 3a, 3b are arranged diagonally in one lane of the road 100 (see FIG. 10). Is different. In this example, as shown in FIG. 9 (A), the elements 3a and 3b are staggered in the vertical direction, that is, shifted diagonally, and the position of the thermopile element with respect to the infrared transmission lens is adjusted as appropriate, and the vehicle detection system is When installed, the monitoring ranges formed by the respective elements are arranged side by side in the vehicle width direction and the lane direction. Note that the monitoring range formed by each element can be shifted and aligned in the lane direction and the lane direction only by adjusting the position of the thermopile element with respect to the central axis of the infrared transmission lens.

  By arranging the two monitoring ranges also in the vehicle width direction, the system 1b can, for example, monitor the range 3B formed based on the engine unit of the vehicle 200, the element 3b in the monitoring range 3A formed based on the thermopile element 3a. As in the vehicle roof, different parts of one vehicle can be monitored by different monitoring ranges 3A and 3B, that is, detected by different elements 3a and 3b.

  As described above, the system 1b includes a plurality of thermopile elements and provides a plurality of monitoring ranges for one lane in one area where the detection of the vehicle is desired. I can pass. In other words, by detecting infrared rays from each part of the vehicle in each monitoring range with different thermopile elements, even when the temperature difference between a certain part of the vehicle and the road surface hardly occurs, The possibility that a temperature difference between the part and the road surface occurs can be increased. In particular, if at least one monitoring range is provided so that a part having a relatively high temperature in the vehicle, for example, an engine unit, a tire, or the like passes, it is less affected by water droplets even in rainy weather, and the vehicle is more accurate. It is possible to increase the possibility of being detected.

  Therefore, the detection system 1b according to the present invention can achieve the excellent effects of the first embodiment and the effect of arranging the above-described monitoring ranges in the vehicle width direction. In other words, in this system, by providing two monitoring ranges in the vehicle width direction, the possibility of causing a temperature difference between the vehicle and the road surface in at least one of the monitoring ranges is improved, thereby affecting the environment such as temperature and weather. The vehicle can be detected with high accuracy. In addition, this system provides two monitoring ranges in the lane direction, which increases the possibility that the vehicle will stop in at least one of the monitoring ranges in a traffic jam. It is possible to detect a heavy traffic situation. Therefore, when the vehicle detection system of the present invention shown in this example is used, cases where it is difficult to detect the vehicle can be further reduced.

<Algorithm for vehicle presence determination 1>
Next, a vehicle detection operation procedure for determining whether or not a vehicle is present when the vehicle detection system shown in the first and second embodiments is constructed will be described in detail. FIG. 11 is a flowchart illustrating an example of a processing procedure in the vehicle presence / absence determining unit. In this example, the vehicle presence / absence determining unit starts processing in accordance with the operation of the sensor unit (step 20). Immediately after the start of processing, initial learning of the background level and threshold is performed (step 21). Depending on the CPU, it may be difficult to determine only the background data from the data when the vehicle is running. Therefore, it is preferable not to determine the presence or absence of the vehicle during the initial learning of the background level. Therefore, it is preferable to turn on the power supply of the sensor unit and the vehicle presence / absence determination unit when there is no vehicle, or to provide a trigger such as a reset. Since the initial learning of the background level is performed when there is no vehicle, it is preferable that the background level be as short as possible, for example, about 1 second. Specific examples of the initial learning of the background level include taking an average value of input level values when no vehicle is present. On the other hand, when the initial learning of the background level is performed when the vehicle exists, for example, the mode value of the input level value for a certain time may be taken.

  Then, after the initial learning of the background level is completed, the vehicle presence / absence determination unit learns a threshold value for a certain time (for example, about 10 seconds). If the initial value is sufficiently large, the threshold value is automatically lowered to an appropriate value by learning. By performing the initial learning in such a procedure, it is possible to take a normal value for both the background level and the threshold value, that is, a value more suitable for the actual environment. The obtained background level and threshold value are stored in a memory.

  After the initial learning, the vehicle presence / absence determination unit starts determining whether or not there is a vehicle. First, the electromotive force obtained from each thermopile element of the sensor unit is amplified and sent to the vehicle presence / absence determination unit to obtain an input level value (step 22). Based on these input level values, a comparison value is calculated in the procedure described later (step 23). Further, a threshold value is calculated based on these input level values in the procedure described later (step 24). Then, the comparison value calculated based on these input level values is compared with the threshold value (step 25). If the comparison value is greater than or equal to the threshold value, it is determined that there is a vehicle (step 26), and “vehicles are present” is written in the sensing and counting result and stored in memory (step 27). In this example, each input level value is also stored (step 29) and used for the background level calculation described later.

  On the other hand, if the comparison value is less than the threshold value, the vehicle presence / absence determining unit determines that there is no vehicle (step 28), and similarly writes “no vehicle” in the sensing count result and stores it in the memory (step 27). At this time, any thermopile element has detected a background such as a road. Each input level value used for the determination is stored for use in calculation (step 29), and the background level calculation means calculates the background level using these input level values (step 30).

  The system of this example includes a wireless communication unit that wirelessly transmits the result of sensing and counting to a signal controller or a management center (see FIG. 2). The wireless communication unit appropriately checks whether or not the sensing and counting result obtained in step 27 is stored in the memory, and sends the counting result to a signal controller or the like according to the procedure described later. In addition, when the power supply to the wireless communication unit is intermittently performed, transmission of the aggregation result to the signal controller or the like is performed when the power of the transmission / reception unit of the wireless communication unit is ON. When the power of the transmission / reception unit is OFF, save it as it is and send it after it is turned ON.

Next, background levels used in the system shown in this example will be described. In this example, an arithmetic value by exponential smoothing is used as the background level. The background level is separately determined by the background level calculation means based on the input level value obtained from each thermopile element. For example, in the example shown in FIGS. 3 and 9, the background level [1] is calculated based on the input level value [1] from the thermopile element 3a, and the background level [2] is calculated based on the input level value [2] from the element 3b. 2] to obtain two background levels [1] and [2]. As a specific calculation of the background level, when the input level value is b _n , the background level used for the next determination is a _n , and the smoothing coefficient is α, a _n = a _n−1 + α × (b _n −a _n−1 ) (a _n−1 is the background level used in the previous vehicle determination). Further, in this example, the smoothing coefficient α is changed according to the previous vehicle determination result so that the background level is more appropriate to the actual environment without depending on the vehicle temperature. Specifically, when the previous determination result is with a vehicle, the smoothing coefficient α is set to a smaller value (α _ON ), for example, 0, and when the previous determination result is no vehicle, the smoothing coefficient α is relatively large. Value (α _OFF ), for example, 0.025. By using the background level by the exponential smoothing method, a more accurate value corresponding to the actual environment can be obtained, and the presence / absence of the vehicle can be more accurately determined.

A procedure for calculating the background level will be described. FIG. 12 is a flowchart showing a background level calculation procedure. First, the vehicle presence / absence determination unit calls the previous vehicle determination result from the memory, and checks whether the determination result is a vehicle (step 40). If the previous determination is that there is a vehicle, α _ON is selected as the smoothing coefficient (step 41). When the previous determination is that there is no vehicle, α _OFF is selected as the smoothing coefficient (step 42). Then, the background level calculation means calculates the input level value b _n , the previous background level an _n−1 called from the memory, and the selected smoothing coefficient as the background level arithmetic expression a _n = a _n−1 + α × (b _n− a _n−1 ) to calculate the background level (step 43). This series of procedures is performed for each input level value to obtain a background level based on each input level value.

Next, comparison values used in the system shown in this example will be described. In this example, a value based on the difference between the background level obtained by the above calculation and the input level value is calculated, and this calculated value is used as a comparison value. Specifically, the vehicle presence / absence determination unit obtains an absolute value (referred to as a background difference) of a difference between the input level value b _n and the previous background level an _n−1 for each input level value. In the example shown in FIGS. 3 and 9, the absolute value of the difference between the input level value [1] from the thermopile element 3a and the background level [1], the input level value [2] from the element 3b and the background level [2] Obtain two background differences [1] and [2], which are the absolute values of the differences. Next, the sum of the obtained background differences (this is called the difference sum) is taken (in the above example, background difference [1] + background difference [2]), and the integrated value of this difference sum over a certain period of time is used as a comparison value. And The absolute value of the difference between the input level value and the background level is obtained for each input level value, and the sum of these background differences is taken to increase the possibility that there is a temperature change in at least one monitoring range, thereby detecting the vehicle. In addition, even if the temperature change is small in all the monitoring ranges, the change can be increased and the vehicle can be detected more reliably. In addition, by using the integrated value of the difference sum as a comparison value, it is possible to reduce cases where the vehicle is misidentified or cannot be recognized even in a traffic jam.

A procedure for calculating the comparison value will be described. FIG. 13 is a flowchart showing a comparison value calculation procedure, and shows a case where an integrated value is used as a comparison value. First, the vehicle presence / absence determining unit calculates a difference (background difference) between the input level value and the background level in each input level value (step 50). In this example, the background difference S _n is the absolute value of the difference between b _n and a _n−1 , that is, S _n = | b _n −a _n−1 |. In the example shown in FIGS. 3 and 9, two background differences _{Sn [1]} and _{Sn [2]} are obtained. Next, the sum (difference sum) of the obtained background differences is taken (step 51). In the above example, the difference sum SS _n = S _{n [1]} + S _{n [2]} is calculated. Then, an integrated value I _n = ΣSS _k obtained by integrating the difference sum SS _n for a predetermined time is calculated (step 52). In this example, each thermopile element obtains an input level value every predetermined time, for example, every 10 ms, and obtains a background difference, respectively, for a certain time, for example, 160 ms difference difference, that is, 16 difference sums. Is accumulated. In the present example, this manner is used as a comparison value integrated value I _n obtained. In addition, you may change suitably integration time, the frequency | count of integration, etc. The same applies to the case where the amount of change per unit time of the integrated value and each input level value described later is used as the comparison value.

  Another comparison value will be described. The above integrated value alone may take 0, and there is a risk of misidentifying the vehicle at this time. Therefore, it is preferable to use a calculation value using the integrated value and the amount of change per unit time of each input level value as the comparison value. Specifically, the vehicle presence / absence determination unit obtains the absolute value (background difference) of the difference between the input level value and the previous background level for each input level value, and takes the sum (difference sum). Then, an integrated value for a certain time of the difference sum is obtained. Next, a change amount is obtained for each input level value, and an average value of the obtained change amounts is obtained, and a sum (hereinafter referred to as a change amount sum) is obtained. In the example shown in FIGS. 3 and 9, two variations [1], that is, a variation [1] of the input level value [1] from the thermopile element 3a and a variation [2] of the input level value [2] from the element 3b ] And [2] are obtained, and the sum of the average value of the change amount [1] and the average value of the change amount [2] is obtained. A value obtained by multiplying the change amount sum by a constant and the integrated value are added as a comparison value. In this way, the value calculated using not only the integrated value but also the amount of change is used as the comparison value, thereby further reducing the case where the vehicle is misidentified or cannot be recognized. In addition to using a thermopile element with a large output and using a value that takes into account the amount of change in this way into the algorithm, it is possible to sufficiently detect infrared rays using only the thermopile element without using another sensor such as a pyroelectric sensor. To determine the presence or absence of the vehicle.

A procedure for calculating the comparison value will be described. FIG. 14 is a flowchart showing the calculation procedure of the comparison value, and shows a case where the integrated value and the change amount are used as the comparison value. First, similarly to FIG. 13, the vehicle presence / absence determining unit calculates a difference between the input level value and the background level, that is, a background difference S _n = | b _n −a _n−1 | 50). Next, the sum (difference sum) SS _n of the obtained background differences is calculated (step 51). Furthermore, it calculates the integrated value I _n = ΣSS _k obtained by integrating the difference sum SS _n predetermined time (step 52). In this example, at each input level value, it integrates the difference sum SS _n of the same manner as described above 16 times. Next, for each input level value, a change amount D _n per unit time of the input level value is calculated (step 53). In this example, the difference is not the difference from the previous input level value but the difference from the previous input level value. Specifically, for example, when the input level value is acquired every 10 ms, the absolute value | b _n −b _n−16 | of the difference from the input level value b _n−16 before 160 ms is set as the change amount D _n . . In the example shown in FIG. 3 and 9, the change amount D _{n [1]} of the input level value from the thermopile element 3a [1], calculates the input level value from the element 3b [2] change amount D _{n [2]} of . Then, by calculating the average value of the change amount obtained respectively sums to determine the amount of change sum IS _n (step 54). In this example, the average of the change amount for a specific number, for example, 8 is taken. That is, the average value B _n = 1/8 × ΣD _k (where k = n−7 to n). In the above example, the average value B _n [1] for the input level value [1] from the thermopile element 3a = 1/8 × ΣD _{k [1]} and the average value B _n for the input level value [2] from the element 3b _[2] = 1/8 × ΣD _{k [2]} is obtained, and these average values B _{n [1]} and B _{n [2]} are added to obtain the sum of changes IS _n = B _{n [1]} + B _{n [2]} Is calculated. Then, by computing the sum of a constant multiple of the integrated value I _n variation sum it IS _n, obtains a comparison value (step 55). That is, in this example, the comparison value is obtained from I _n + IS _n × A (where A is a constant, for example, A = 0 to 255, initial value 30). Constant A is an arbitrary value determined integrated value I _n, the variation sum IS _n to depending to what degree involved respectively. Note that the calculation value of each step is stored in a memory so that it can be used for the next calculation. Further, the specific number for obtaining the average value and the constant A may be appropriately changed.

Next, threshold values used in the system shown in this example will be described. The threshold is also preferably changed according to the actual environment. In this example, the threshold value is varied depending on the change in the amount of infrared rays, that is, the temperature change. Here, the magnitude of the temperature change tends to occur with the previous vehicle determination result. Therefore, in this example, the threshold value is changed according to the previous vehicle determination result. Specifically, the setting value is set to the lowest value, and a correction value is added to the setting value to follow the environment. In this example, this correction value is changed according to the previous vehicle determination result. In this example, the correction value is obtained by adding a value based on the average value of the integrated values to the previous correction value. Specifically, this correction value = previous correction value + {(average value of integrated values × β) −previous correction value} × θ (where β is a correction coefficient, θ is a constant, for example, 0.01 Do). Then, β is changed in order to change the correction value according to the previous vehicle determination result. More specifically, when the previous determination result is with a vehicle, the correction coefficient β is set to a smaller value (β _ON ), for example, 12. When the previous determination result is no vehicle, the correction coefficient β is relatively large value (beta _OFF), for example to 36. Thus, by changing the correction value and changing the threshold value, once it is determined that the vehicle is present, it is easy to continue the determination that the vehicle is present. Further, when the vehicle is stopped due to traffic jams or the like, it can be easily determined that the vehicle is present, and when the stopped vehicle starts traveling, it can be easily determined that there is no vehicle. In addition, when calculating | requiring a threshold value by calculation in this way, a threshold value may become large depending on the case. Therefore, it is preferable to provide a maximum value for the threshold value.

A procedure for calculating this threshold will be described. FIG. 15 is a flowchart illustrating a threshold calculation procedure. First, the vehicle presence / absence determining unit calculates an average value of integrated values at each input level value (step 60). In this embodiment, for example, a 1/16 × I _n. Therefore, for example, when the input level value is acquired every 10 ms in the same manner as described above, the background difference is obtained for each 10 ms for each input level value and summed, and the difference sum for 160 ms, that is, the difference sum for 16 times is obtained. used divided by the integrated value I _n the 16 obtained by integrating. It may be used to call the integrated value I _n stored in the memory. Next, the previous vehicle determination result is called from the memory in the same manner as in the calculation of the background level, and it is confirmed whether or not the determination result is that there is a vehicle (step 61). If the previous determination is that there is a vehicle, β _ON (β _{L in} FIG. 15) is selected as the correction coefficient (step 62). A correction value C _L is calculated using this correction coefficient β _ON (step 63). That is, C _L = previous correction value + {(1/16 × I _n × β _ON ) −previous correction value} × 0.01 (assuming θ = 0.01). Then, the correction value C _L is added to a set value (for example, 1000) to calculate a threshold value L ′ (step 64). Further, the threshold L is calculated by multiplying the threshold L ′ by a hysteresis coefficient (step 65).

  In this example, as the threshold value L, hysteresis coefficient × [setting value + [previous correction value + {(average integrated value × β) −previous correction value} × θ]] is used. By applying a hysteresis coefficient only to the threshold value L, it is easy to make a difference from the threshold value H, which will be described later, and when the comparison value fluctuates near the threshold value, the determination of the presence of the vehicle and the determination of the absence of the vehicle occur repeatedly in a very short time. Such an excessive determination can be prevented. The hysteresis coefficient is, for example, 0.7 when the previous determination result is with a vehicle and 1.0 when there is no vehicle.

On the other hand, when the previous determination is that there is no vehicle, the vehicle presence / absence determination unit selects β _OFF (β _{H in} FIG. 15) as the correction coefficient (step 66). A correction value _CH is calculated using this correction coefficient β _OFF (step 67). That, C _H = last correction value + - (a theta = 0.01) obtaining a _{{(1/16 × I n} × β OFF) the last correction value} × 0.01. Then, by adding the correction value C _H a set value (e.g. 1000), calculates the threshold H (step 68).

  The correction coefficient β, the specific number for obtaining the average value, and the constant θ may be changed as appropriate. In particular, when only the integrated value is used as the comparison value and when the integrated value and the change amount are used, the threshold value may be changed by changing these values, or the threshold value may be changed by another calculation. Also good. In addition, the calculation value of each step is stored in a memory so that it can be used for the next calculation.

<Algorithm for detecting vehicle identity>
The vehicle detection system shown in the first and second embodiments (see FIGS. 3 and 9) is easy to detect a vehicle even in a time zone where the vehicle is difficult to detect, and has the effect of improving the accuracy of occupancy measurement in a traffic jam. In addition, by providing the identity determination unit, it is possible to reduce the risk of misidentifying one vehicle as two or more.

  The identity determination unit determines the identity of the vehicle using the output waveform of the input level value obtained from both thermopile elements. Specifically, the identity determination unit corresponds to the first waveform from the time difference between the first waveform and the second waveform adjacent to each other in any output waveform and the speed calculated by the vehicle detector. The inter-vehicle distance calculating means for calculating the distance between one vehicle and the second vehicle corresponding to the second waveform is compared with the calculated inter-vehicle distance and the set value, and whether or not the first vehicle and the second vehicle are the same Identity determining means for determining whether or not. Hereinafter, an operation procedure for reducing the misidentification of the number of vehicles will be described.

  FIG. 16 is a flowchart showing an operation procedure for determining whether or not the input level obtained from the thermopile element is that of the same vehicle. In the input level value from the thermopile element, when a waveform appears continuously as shown in FIG. 1, two adjacent waveforms (here, waveform A (first waveform), waveform B (second waveform)), or two It may be difficult to distinguish whether two or more waveforms are from the same vehicle or from another vehicle. In such a case, in the vehicle detection system of the present invention in which a plurality of monitoring ranges (two in FIG. 4) can be provided in the lane direction as shown in FIG. 4, the vehicle speed is obtained from the input level value by the vehicle detector. In addition, by providing the identity determination unit, it is possible to determine the distance between the vehicles that draw the waveform, and determine whether the continuous waveform appearing in the input level value is due to the same vehicle.

  Specifically, first, as shown in FIG. 4, the distance x between the two monitoring ranges 3A and 3B provided in the lane direction is obtained in advance. The distance x can be determined using, for example, the installation position of the vehicle detection system 1a and the difference in the directivity angle between the two thermopile elements. This distance x is stored in the memory as a set value. In addition, a limit value of the distance between vehicles traveling at a specific speed is set in advance and stored in the memory. For example, when traveling at a speed of 70 km / h, the limit value of the distance between vehicles is set to 5 m.

  Next, the vehicle detector calculates a time difference t between the output waveforms 3A ′ and 3B ′ based on each thermopile element (step 110). More precisely, a correlation function is calculated from waveform A (waveform B) of both output waveforms 3A ′ and 3B ′, and the maximum value of the correlation function is obtained. Next, the vehicle speed v = x / t is calculated from the distance x called from the memory and the time difference t obtained by the calculation (step 111). Then, the inter-vehicle distance calculation means obtains a time ta between the waveform A and the waveform B from the output waveform A ′ (output waveform B ′), and draws the vehicle and the waveform B that draw the waveform A from the speed v and the time ta. A distance d between the vehicles is calculated (step 112). Next, the identity determination means calls the limit value of the distance between the vehicles with respect to the speed v from the memory, and compares the calculated distance d with the called limit value (step 113). If the distance d is greater than or equal to the limit value, the identity determination means determines that the vehicle that draws the waveform A and the vehicle that draws the waveform B are different vehicles (step 114). On the other hand, when the distance d is less than the limit value, the identity determination means determines that the vehicle drawing the waveform A and the vehicle drawing the waveform B are the same vehicle (step 115). These procedures may be performed after, for example, obtaining the input level value (steps 22 and 92) in the operation procedure shown in FIG. 11 and FIG. 19 described later, for example. Specifically, it may be performed after the vehicle presence / absence determination unit determines that there is a vehicle.

  Next, wireless communication means used in the system shown in this example will be described. The radio communication unit 11 shown in FIG. 2 includes a radio control unit 11b and a transmission / reception unit 11a having an antenna, etc., and the aggregation results sent from the vehicle presence / absence determination unit 10 are used as a signal controller, a management center, etc. To send to. In this example, a power supply control unit using a timer (not shown) intermittently supplies power to the transmission / reception unit 11b, and power is input to the transmission / reception unit 11a only for a certain period of time, and transmission / reception of the counting results is performed. Do. FIG. 17 is a flowchart showing a control procedure of the timer, and FIG. 18 is a flowchart showing a processing procedure in the radio control unit.

  As shown in FIG. 17, for example, the timer counts up with a period of 10 ms (step 70), and controls ON / OFF of the power of the transmission / reception unit at a period of 10 s. In the 10s period, the power of the transmitter / receiver is turned on for 500ms from 0 to 500ms (step 71), and the sensing process result is transmitted / received. When the time is 500ms, the power of the transmitter / receiver is turned off (step 72). Stop. Further, the processing of the wireless control unit is started at 100 ms (step 73). Then, when 10s is reached, the reset is performed (step 74), the count-up is newly started, and this is repeated. In this example, the processing start time (100 ms) of the radio control unit is delayed by 100 ms from the operation start time (0 ms) of the transmission / reception unit. Further, the count-up cycle, the power ON start time and operation time of the transmission / reception unit, the reset cycle, and the like may be changed as appropriate.

  When the power of the transmission / reception unit is turned on, processing of the radio control unit starts as shown in FIG. 18 (step 80). The wireless control unit reads the result of sensing and counting received from the vehicle presence / absence determination unit (step 81), and transmits the result read via the transmission / reception unit such as a signal controller or a management center (step 82). After the transmission, the sensing total result stored in the memory is cleared (step 83). It should be noted that it is also possible to accumulate the sensing count results without clearing them and send only the latest sensing count results.

<Algorithm for vehicle presence determination 2>
Another example of the vehicle detection operation procedure for determining whether or not a vehicle is present when the vehicle detection system of the present invention is constructed will be described. FIG. 19 is a flowchart showing another example of the processing procedure in the vehicle presence / absence determining unit, and FIG. 20 is a flowchart showing the procedure of comprehensive determination. The vehicle presence / absence determination unit shown in this example includes a first determination unit and a second determination unit, and by using these determination units, the sum of the input level values obtained from each thermopile element and the difference value of the vehicle are used. This is a configuration for performing multi-stage determination in which presence / absence determination (hereinafter referred to as presence determination) is performed, and finally determination (hereinafter referred to as comprehensive determination) is performed based on the result of the presence determination. In this example, the vehicle presence / absence determination unit starts processing in accordance with the operation of the sensor unit as in the algorithm 1 (step 90). Immediately after the start of processing, initial learning of the background level and threshold is performed in the same procedure as in algorithm 1 (step 91), and then the presence / absence of a vehicle is determined.

  In order to determine the presence or absence of a vehicle, first, presence determination is performed. Specifically, the input level value is obtained from each thermopile element of the sensor unit as in algorithm 1 (step 92). The sum and difference of the obtained input level values are taken, and the sum value of the input level values and the difference value of the input level values are respectively calculated (step 93). The calculation of the difference includes taking the absolute value of the difference between the input level values.

  Next, the vehicle presence / absence determination unit determines a value based on the difference between the obtained input level value sum and the background level sum value, and a difference between the input level value difference value and the background level difference value. Are respectively compared values (first comparison value and second comparison value), and these comparison values are respectively calculated according to the procedure described later (step 94). Further, based on the sum value and the difference value, thresholds (first threshold value and second threshold value) for each are calculated in the procedure described later (step 95).

  Next, the first determination means of the vehicle presence / absence determination unit compares the first comparison value calculated based on the sum value with the first threshold value, and performs presence determination (step 96). Similarly, the second determination unit compares the second comparison value calculated based on the difference value with the second threshold value. In each comparison, if the comparison value is equal to or greater than the threshold value, it is determined that there is a vehicle (step 97), and the result is temporarily stored in the memory (step 98). The vehicle presence result based on the sum of the input level values is defined as a first presence result, and the vehicle presence result based on the difference value of the input level values is defined as a second presence result.

  On the other hand, in each comparison, when the comparison value is less than the threshold value, the first determination means and the second determination means determine that there is no vehicle (step 99), and similarly the results (first existence result, second existence result). ) Is temporarily stored in the memory (step 98). Also, the input level value used for existence determination is stored (step 100) and used for the background level calculation (step 101). The presence determination is completed by the above procedure.

  Next, the comprehensive determination means of the vehicle presence / absence determination unit performs the comprehensive determination shown in the flowchart of FIG. Specifically, the sum (or) of the obtained first existence result and second existence result is taken (step 102). The presence or absence of a vehicle is determined based on this or result, and a final determination is made (step 103). When both the first presence result and the second presence result have a vehicle, the or result is a vehicle. Therefore, it is finally determined that there is a vehicle (step 104). When one of the first presence result and the second presence result is with a vehicle and the other is without a vehicle, the or result is “with vehicle”. Therefore, it is finally determined that there is a vehicle (step 104). On the other hand, when both the first presence result and the second presence result are without a vehicle, the or result is no vehicle. Therefore, it is finally determined that there is no vehicle (step 105). The result of comprehensive judgment is written in the result of sensing and counting and stored in the memory (step 106). In this example, the sum value and difference value used for presence determination are also stored (step 107) and used for comparison value calculation described later. The obtained sensor count result is transmitted to the signal controller and the management center by the wireless communication unit.

As the background level used in the system shown in this example, an arithmetic value obtained by exponential smoothing may be used as in algorithm 1. That is, the background level is obtained and used separately based on the input level value obtained from each thermopile element. For example, in the example shown in FIGS. 3 and 9, the background level [1] is calculated based on the input level value [1] from the thermopile element 3a, and the background level [2] is calculated based on the input level value [2] from the element 3b. 2] to obtain two background levels [1] and [2]. As a specific calculation of the background level, when the input level value is b _n , the background level used for the next presence determination is a _n , and the smoothing coefficient is α, a _n = a _n−1 + α × (b−a _n−1 ) (a _n−1 is the background level used for the previous existence determination). If the smoothing coefficient is changed according to the result of the previous presence determination as in Algorithm 1, a more accurate value can be obtained according to the actual environment, and a more accurate vehicle presence determination can be performed. It is preferable.

The procedure for calculating the background level may be the same as in Algorithm 1. That is, first, the result of the previous existence determination is called from the memory, and it is confirmed whether the determination result is whether there is a vehicle. If there is a vehicle, the smoothing coefficient is selected as α _ON, and if there is no vehicle, the smoothing coefficient is , Α _OFF is selected. Then, the input level value b _n , the previous background level a _n−1 called from the memory, and the selected smoothing coefficient are calculated as the above background level arithmetic expression a _n = a _n−1 + α × (b _n −a _n−1 ) To calculate the background level for each input level value. In this example, the sum value and the difference value of the background levels obtained by the above calculation are used for the calculation of the comparison value described later. For example, in the example shown in FIGS. 3 and 9, the input level value [1] from the thermopile element 3a, the background level [1] based on the input level value [1], the input level value [2] from the element 3b, and The background level [2] based on the input level value [2] is obtained. At this time, the sum of the input level values is the input level value [1] + input level value [2], and the sum of the background levels is the background level [1] + background level [2]. The value may be input level value [1] −input level value [2], and the background level difference value may be background level [1] −background level [2].

  Next, comparison values used in the system shown in this example will be described. In this example, the value calculated based on the difference between the sum of the input level value and the background level is the first comparison value, and the value calculated based on the difference between the input level value difference value and the background level difference value. Is the second comparison value. Specifically, the difference (called background difference) between the sum value (difference value) of the input level value and the previous background level sum value (difference value) is obtained, and the integrated value of this background difference over a certain period of time is calculated. One comparison value (second comparison value) is used. In this way, the calculation value using the difference between the sum value (difference value) of the input level value and the sum value (difference value) of the background level is used as a comparison value, thereby reducing cases where the vehicle is misidentified or cannot be recognized. To do.

A procedure for calculating the comparison value will be described. First, a case where an integrated value is used as a comparison value will be described. Difference S _{n [1]} between the sum of the input level values and the previous background level (background difference [1]), and the difference S _{n [} between the difference between the input level values and the previous background level difference _{[ 2]} (Background difference [2]) is calculated. In this example, similarly to Algorithm 1, the absolute value of the difference between the sum value (difference value) of the input level values and the previous sum value (sum value) of the background level is taken. For example, in the examples shown in FIGS. 3 and 9, the background difference [1] is set to | (input level value [1] + input level value [2]) − (background level [1] + background level [2]) | The difference [2] may be | (input level value [1] −input level value [2]) − (background level [1] −background level [2]) |. Then, similarly to Algorithm 1, the background subtraction [1], the integrated value obtained by integrating a certain period of time [2] [1]: I n [1] = ΣS k [1], the integrated value [2]: I _{n [2]} = ΣS _{k [2]} is calculated. In this example, using the integrated value I _n obtained in this way in the first comparative value, as a second comparison value. In addition, you may change suitably integration time, the frequency | count of integration, etc. The same applies to the case where an integrated value and a change amount described later are used as comparison values.

  Next, a case where a change amount is used as a comparison value will be described. Specifically, in the same manner as described above, the difference between the sum value (difference value) of the input level values and the sum value (difference value) of the previous background level, that is, the background differences [1] and [2] are respectively obtained. Then, the integrated values [1] and [2] of the background differences [1] and [2] over a fixed time are obtained. Next, a change amount of the sum value (difference value) of the input level values is obtained, and an average value of the obtained change amounts is obtained. A value obtained by multiplying the amount of change by a constant and the integrated value are added to obtain a comparison value. In this way, the value calculated using not only the integrated value but also the amount of change is used as the comparison value, thereby further reducing the case where the vehicle is misidentified or cannot be recognized. In addition to using a thermopile element with a large output and using a value that takes into account the amount of change in this way into the algorithm, it is possible to sufficiently detect infrared rays using only the thermopile element without using another sensor such as a pyroelectric sensor. To determine the presence or absence of the vehicle.

A procedure for calculating the comparison value will be described. Similar to the case where the integrated value [1] and integrated value [2] are used as the first comparison value and the second comparison value, the sum value (difference value) of the input level value and the sum value (difference value) of the background level. Difference, background difference [1], background difference [2] are obtained, and these background differences [1] and [2] are integrated for a certain period of time [1]: In _[1] , integrated value [2]: I Calculate _{n [2]} . Next, a change amount [1]: D _{n [1]} and a change amount [2]: D _{n [2]} per unit time are calculated for the sum value and the difference value of the input level values. For example, the absolute value of the difference from the sum (difference value) of the input level value before 160 ms is set as the change amount D _n . Next, the average value B _{n [1] of} the obtained variation D _{n [1] and} the average value B _{n [2]} of the variation D _{n [2]} are calculated. For example, an average value of eight (8) the amount of change _{B n = 1/8 × ΣD} k ( where, k = n-7~n). Then, the integrated value I _{n [1]} and the average value sum I _{n [1]} of the integral multiples of _{B n [1] + B n} [1] × A ( where, A is a constant, for example, A = 0 to 255, the initial The value 30) is calculated to obtain a first comparison value based on the sum of the input level values. Also, the sum of the integrated value In _[2] and a constant multiple of the average value _{Bn [2] In [2]} + _{Bn [2]} × A (where A is a constant, for example, A = 0 to 255, initial The value 30) is calculated to obtain a second comparison value based on the difference value of the input level values. Constant A, the integrated value I _n, is any value determined by the average value B _n whether to what extent involved respectively. Note that the calculation value of each step is stored in a memory so that it can be used for the next calculation. Further, the specific number for obtaining the average value and the constant A may be appropriately changed.

  The threshold values used in the system shown in this example are provided for the sum value and the difference value of the input level values, and arithmetic values are used so as to change according to the actual environment as in the algorithm 1. Specifically, this correction value = previous correction value + {(average value of integrated values × β) −previous correction value} × θ (where β is a correction coefficient, θ is a constant, for example, θ = 0.01. ) And β is changed according to the result of the previous existence determination. By changing the threshold value in this way, it is easy to continue the presence determination that the vehicle is present while the vehicle is stopped, and to make it easy to determine the presence of the vehicle when the stopped vehicle starts traveling. it can. Note that a maximum value is preferably set as the threshold value.

The procedure for calculating this threshold value may be the same as in Algorithm 1. That is, first, the integrated values In _[1] and In _[2] of the sum and difference values of the input level values are obtained, and the average value of these integrated values In _[1] and In _[2] is obtained. Calculate each. For example, the average of the integrated values I _n of the sum value of the input level value (difference value) and 1/16 × I _n. May be used to call the integrated value I _n of the stored in the memory sum value (difference value). Next, as in the case of the background level calculation, the result of the previous presence determination is called from the memory to check whether the presence determination result indicates that there is a vehicle. If there is a vehicle, select β _ON as the correction coefficient. . The correction value C _L = last correction value + sum value of the input level value with a correction coefficient beta _ON for (difference value) {(1/16 × I n × β ON) - the last correction value} × 0.01 ( θ = 0.01) is calculated, the correction value C _L is added to a set value (for example, 1000), and a threshold value L ′ for the sum value (difference value) of the input level values is calculated. Further, a threshold value L is obtained by multiplying the threshold value L ′ by a hysteresis coefficient.

On the other hand, if the previous presence determination is no vehicle, select β _OFF as the correction coefficient, and use this correction coefficient β _OFF to correct the correction value C _H for the sum value (difference value) of the input level value = previous correction Value + {(1/16 × I _n × β _OFF ) −previous correction value} × 0.01 (θ = 0.01) is calculated. The correction value C _H is added to a set value (for example, 1000), and a threshold value H for the sum value (difference value) of the input level values is calculated. The correction coefficient β, the specific number for obtaining the average value, and the constant θ may be changed as appropriate. In particular, when only the integrated value is used as the comparison value and when the integrated value and the change amount are used, the threshold value may be changed by changing these values, or the threshold value may be changed by another calculation. Also good.

<Circuit configuration>
In the above example, it is described that a thermopile element having a large output is used and a value used for the algorithm is set as a specific calculation value so that the vehicle can be detected accurately even using only the thermopile element. In addition, a circuit configuration can be devised as a configuration capable of detecting a vehicle even in a traffic jam. FIG. 21 is a circuit diagram schematically showing a connection state between the thermopile element and the CPU in the vehicle detection system of the present invention. As described above, the circuit includes the thermopile elements 3a, 3b and the vehicle presence / absence determination unit 10 having a CPU, and amplifies the electromotive force generated in each element 3a, 3b to the determination unit 10. It includes amplifiers 13a and 13b for sending. In this example, as shown in FIG. 21, each of the elements 3a and 3b includes two amplifiers 13a and 13b, and the amplifier 13a connected to each of the elements 3a and 3b has a large amplification factor (FIG. 21). The amplifier 13b connected to the CPU is assumed to have a small amplification factor (10 times the same). The amplifier 13b is configured to amplify the difference between the output of the amplifier 13a and the reference voltage by adding the reference voltage by connecting to the D / A converter 15. With this configuration, the dynamic range of the CPU can be improved and processing can be performed appropriately. That is, the input level value obtained by amplifying the electromotive force generated in each element 3a, 3b by the amplifiers 13a, 13b can be used for vehicle detection, vehicle speed calculation, and vehicle identity determination.

  Note that the reference voltage to be applied may be adjusted by appropriately examining the potential of the amplifier 13b. In this example, the reference voltage is adjusted by the CPU. In the example shown in FIG. 21, one CPU is provided for the two thermopile elements 3a and 3b. However, one CPU may be provided for each element 3a and 3b. Thus, by amplifying not only the AC component but also the DC component, it is possible to continue to detect a stopped vehicle. Therefore, even when the vehicle is stopped due to traffic jam or stagnation, the vehicle can be detected with high accuracy.

  The vehicle detector of the present invention can be preferably used when the speed of a vehicle traveling on a road is obtained. In addition, the vehicle detection system of the present invention including this detector can be suitably used for detecting the presence or absence of a vehicle on the road and detecting the speed of the vehicle.

It is a model figure which shows the waveform of the input level value obtained from the thermopile element. It is a functional block diagram of the vehicle detection system of the present invention. (A) is the front view which showed typically the vehicle detection system of this invention which forms two monitoring ranges in a lane direction, (B) is the side surface BB sectional drawing. It is explanatory drawing which shows the state which has arrange | positioned and arrange | positioned the vehicle detection system of this invention which forms two monitoring ranges in a lane direction to the support | pillar of a roadside. It is an expansion schematic diagram which shows the cross section of the infrared transmission lens part in this invention vehicle detection system. It is a schematic diagram for explaining the arrangement state of the thermopile element and the infrared transmission lens, in which each lens is provided with a lens, (A) is an example of shifting the position of the element with respect to the lens, (B) is This is an example in which the position of the support portion is shifted. It is a schematic diagram explaining the arrangement | positioning state of a thermopile element and an infrared rays transmissive lens, and is an example which provides two elements with respect to one lens. It is a perspective view which shows the outline of this invention vehicle detection system. (A) is the front view which showed typically the vehicle detection system of this invention which forms a monitoring range each shifting | deviating to a vehicle width direction and a lane direction, (B) is the side surface BB sectional drawing. It is explanatory drawing which shows the state which has arrange | positioned this invention vehicle detection system which shifts | deviates to a vehicle width direction and a lane direction, and forms two monitoring ranges to the support | pillar of a roadside. It is a flowchart which shows an example of the process sequence in a vehicle presence determination part in this invention vehicle detection system. It is a flowchart which shows the calculation procedure of a background level in the vehicle detection system of this invention. In this invention vehicle detection system, it is a flowchart which shows the calculation procedure of a comparison value, and shows the case where an integrated value is used as a comparison value. In this invention vehicle detection system, it is a flowchart which shows the calculation procedure of a comparison value, and shows the case where an integrated value and variation | change_quantity are used as a comparison value. It is a flowchart which shows the calculation procedure of a threshold value in this invention vehicle detection system. In the vehicle detection system of this invention, it is a flowchart showing the operation procedure which determines whether the input level obtained from the thermopile element is the thing of the same vehicle. 5 is a flowchart illustrating a control procedure of a timer of a wireless communication unit in the vehicle detection system of the present invention. It is a flowchart which shows the process sequence in a radio | wireless communication part in this invention vehicle detection system. It is a flowchart which shows another example of the process sequence in a vehicle presence determination part in this invention vehicle detection system. It is a flowchart which shows the procedure of comprehensive determination in this invention vehicle detection system. FIG. 4 is a circuit diagram schematically showing a connection state between a thermopile element and a CPU in the vehicle detection system of the present invention. It is explanatory drawing which shows the state in which a vehicle detector detects a vehicle at the time of traffic congestion, Comprising: (A) is a monitoring range of a vehicle detector, when (B) exists in the monitoring range of a vehicle detector. Shows the case where there is no vehicle. It is explanatory drawing which shows the state which expanded the monitoring range so that it might become the whole lane in a vehicle detector.

Explanation of symbols

1,1a, 1b Vehicle detection system 2 Sensor part 3,3a, 3b Thermopile element
3A, 3B Monitoring range 4 Enclosure 4a Aiming part 4b Concave protrusion 4c Convex protrusion 4d Hole
5a, 5b substrate 5c CPU substrate 6, 6a, 6b Infrared transmitting lens 6c Lens presser
6d Center shaft 7,7a, 7a ', 7a ", 7b Support 8 Space 9 Screw
10 Vehicle presence / absence determination unit 11 Wireless communication unit 11a Transmission / reception unit 11b Wireless control unit
12 Solar cell part 12a Solar cell 13,13a, 13b Amplifier 14 A / D converter
15 D / A converter
100 Road 200 Vehicle 300 Prop 400 Vehicle detector 401 Monitoring range

Claims (10)

A vehicle detector for detecting the speed of a vehicle existing in a monitoring range on a road,
A first thermopile element and a second thermopile element for sensing infrared rays emitted from a detection target;
A speed detection unit that calculates the speed of the vehicle using the input level values obtained from the two elements;
The both elements are arranged such that the monitoring range based on each element is aligned in the lane direction of one lane,
The speed detector
A time difference calculating means for calculating the time difference of the output waveform of the input level value obtained from each element;
A vehicle detector comprising: a speed calculation means for calculating the speed of the vehicle from the obtained time difference and the distance between the two monitoring ranges. 2. The vehicle detector according to claim 1, wherein the time difference calculating means calculates a correlation function from both output waveforms and calculates the time difference based on the correlation function. The vehicle detector according to claim 1 or 2,
A vehicle presence / absence determination unit that determines the presence / absence of a vehicle present in the monitoring range using the input level values obtained from both thermopile elements;
An identity determination unit for determining the identity of the vehicle,
The identity determination unit
In either of the output waveforms of the input level values obtained from both elements, the first waveform corresponding to the first waveform is calculated from the time difference between the adjacent first waveform and the second waveform and the speed calculated by the vehicle detector. An inter-vehicle distance calculating means for calculating a distance between the one vehicle and the second vehicle corresponding to the second waveform;
A vehicle detection system comprising: an identity determination unit that compares the calculated inter-vehicle distance with a set value to determine whether the first vehicle and the second vehicle are the same. 4. The vehicle detection system according to claim 3, wherein the vehicle detector calculates a time difference and a speed when the vehicle presence / absence determination unit determines that the vehicle is present. In the vehicle presence / absence determination unit,
In the input level value obtained from each thermopile element, a value based on the amount of infrared rays emitted from an object other than the vehicle is used as the background level of each monitoring range, and the background level of each monitoring range is calculated from the value based on this amount of infrared rays. Background level calculation means;
A comprehensive determination means for determining the presence or absence of a vehicle by comparing a comparison value with a threshold value by using a value based on a difference between an input level value of each monitoring range and a background level as a comparison value. The vehicle detection system according to 3 or 4. 6. The vehicle detection system according to claim 5, wherein as the comparison value, a value obtained by integrating a total sum of differences between the input level value and the background level of each monitoring range for a certain period of time is used. 7. The vehicle detection system according to claim 5, wherein a change amount per unit time of each input level value is used as the comparison value. In the vehicle presence / absence determination unit,
In the input level value obtained from each thermopile element, a value based on the amount of infrared rays emitted from an object other than the vehicle is used as the background level of each monitoring range, and the background level of each monitoring range is calculated from the value based on this amount of infrared rays. Background level calculation means;
The sum of the input level values obtained from each thermopile element is taken, the sum of each background level obtained from the background level calculation means is taken, and the difference between the sum of the input level values and the sum of the background levels is calculated. A first determination means for determining a vehicle presence by comparing the first comparison value with a first threshold value based on a value based on the first comparison value;
The difference between the input level values obtained from each thermopile element is taken, the difference between the background levels obtained from the background level calculation means is taken, and the difference between this input level value difference value and the background level difference value is calculated. A second determination unit that determines the presence of the vehicle by comparing the second comparison value with a second threshold value based on the second comparison value;
The comprehensive determination means for determining the presence or absence of a vehicle based on the sum of both the first presence result of the first determination means and the second presence result of the second determination means, or 4. The vehicle detection system according to 4. As the first comparison value, a value obtained by integrating the difference between the sum of the input level values and the background level for a certain period of time is used. As the second comparison value, the difference between the input level value and the background level difference value is used. 9. The vehicle detection system according to claim 8, wherein a value obtained by integrating the difference for a certain time is used. The change amount per unit time in the sum value of the input level values is used as the first comparison value, and the change amount per unit time in the difference value of the input level values is used as the second comparison value. The vehicle detection system according to 8 or 9.

Images