JP7400441B2 - Traffic situation observation device, traffic situation observation method, and traffic situation observation program - Google Patents

Description

The present invention relates to a technique for observing traffic conditions on a road on which a vehicle travels.

BACKGROUND ART Conventionally, in order to observe the traffic conditions of roads (expressways, general roads, etc.) on which vehicles travel, it has been carried out to count the number of vehicles that have traveled through a target area set on the road.

For example, Patent Document 1 describes a device configured to detect a vehicle traveling in a target area by detecting reflected waves of radio waves such as microwaves irradiated to the target area. This Patent Document 1 describes a known Doppler radar (hereinafter sometimes referred to as radio wave radar) that measures the speed of a detected vehicle using the frequency of radio waves irradiated to a target area and the frequency of detected reflected waves. ). In addition, this Patent Document 1 uses the amplitude of reflected waves to separate two vehicles that are continuously running in a target area at approximately the same speed, and erroneously treats the two vehicles as one vehicle. Preventing detection.

Japanese Patent Application Publication No. 2011-204138

However, when detecting a vehicle using radio wave radar, one vehicle may be detected as two vehicles. When one vehicle is detected as being broken into two vehicles, it is referred to here as a detected crack. This detection crack is likely to occur in large vehicles such as trucks and trailers that have a long overall length (vehicle length).

Radio wave radar determines the position of the reflecting surface (vehicle) (the distance from the radio wave irradiation position to the reflecting surface) depending on the radio wave irradiation direction and the time from radio wave irradiation to detecting the reflected wave (radio wave flight time). ) is detected. An event may occur in which radio waves are irradiated onto a connecting part of a vehicle, and the reflected waves cannot be detected by radio radar. The radio radar determines that a vehicle is not present in areas where reflected waves cannot be detected (such as connected areas). Furthermore, the radio wave radar may detect reflected waves that are reflected multiple times due to unevenness of the vehicle body. Since the flight time of a reflected wave that has been reflected multiple times becomes longer, the position of the reflecting surface (vehicle) detected by the radio wave radar will be incorrect (it will be farther away than it actually is). As a result, the radio radar falsely detects that a vehicle is present at a location where it actually does not exist.

Note that it was difficult to determine whether a crack was detected depending on the length of the detected vehicle. Specifically, if one large vehicle is detected as two vehicles, the lengths of the individual detected vehicles will be close to the length of the regular vehicle. For this reason, the vehicle commander cannot properly determine whether the vehicle is one large vehicle or two regular vehicles.

An object of the present invention is to provide a technology that suppresses false detection of a vehicle due to cracked detection and improves the accuracy of observing traffic conditions.

The traffic situation observation device of the present invention is configured as shown below in order to achieve the above object.

The vehicle detection data acquisition unit acquires vehicle detection data including the position and speed of the vehicle detected by scanning a target area set on a vehicle travel route with a probe wave. The exploration wave is, for example, a radio wave such as a microwave or a millimeter wave. Doppler radar (hereinafter referred to as radio wave radar) is a device that detects the position of an object such as a vehicle, the speed of this object, etc. by emitting radio waves such as microwaves and millimeter waves as exploration waves and detecting the reflected waves. ) is publicly known. The vehicle detection data acquisition unit acquires vehicle detection data regarding a vehicle detected by a radio wave radar, for example.

The determination target vehicle extraction section processes the vehicle detection data acquired by the vehicle detection data acquisition section, and extracts two vehicles that are consecutively lined up in the traveling direction on the traveling path. The two vehicles detected in the same lane are vehicles lined up in the direction of travel. In other words, the two vehicles detected in different lanes are not vehicles that were lined up in the direction of travel. When two vehicles are erroneously detected by the radio radar due to a detection error when they are actually one vehicle, the two erroneously detected vehicles are lined up consecutively in the travel direction on the travel path. Therefore, the determination target vehicle extraction unit extracts combinations of vehicles that may have been falsely detected as two vehicles due to detection cracks (combinations of vehicles for which it is necessary to determine whether or not they are false detections), and Combinations of vehicles that are unlikely to have been falsely detected as vehicles (combinations of vehicles that do not require a determination as to whether or not they have been falsely detected) are not extracted.

The detected crack determination unit uses the speeds and inter-vehicle distances of the two vehicles extracted by the determination target vehicle extraction unit to determine whether these two vehicles are detected crack vehicles. The detection crack determination unit sets an inter-vehicle distance determination value according to the speed of the vehicle extracted by the determination target vehicle extraction unit, and determines whether the inter-vehicle distance between the two vehicles extracted by the determination target vehicle extraction unit is greater than the inter-vehicle distance determination value. If it is shorter, it is determined that these two vehicles are one vehicle that has been detected.

Many drivers drive their vehicles while maintaining a distance between them that allows them to avoid a rear-end collision with the preceding vehicle even if the preceding vehicle suddenly decelerates or suddenly stops. From this, it can be determined from the inter-vehicle distance between the two vehicles extracted by the determination target vehicle extracting unit whether one vehicle is erroneously detected due to detection failure. Moreover, since the detected crack determination section sets the inter-vehicle distance determination value used for this determination according to the speed of the vehicle extracted by the determination target vehicle extraction section, determination accuracy can be ensured. It is preferable that the inter-vehicle distance determination value is set according to the speed of the following vehicle extracted by the determination target vehicle extraction unit, but it is preferable to set the inter-vehicle distance determination value according to the speed of the following vehicle extracted by the determination target vehicle extraction unit. It may also be set according to the speed of the preceding vehicle extracted by.

Therefore, erroneous detection of vehicles due to detection cracks can be suppressed, and the accuracy of observing traffic conditions can be improved.

Furthermore, a counting section may be additionally provided to count the number of vehicles that have traveled within the target area based on the determination result of the detected crack determination section. With this configuration, it is possible to prevent the accuracy of counting the number of vehicles from decreasing due to detection cracks. In other words, the accuracy of counting the number of vehicles can be improved.

Further, the detection crack determination section may be configured to set a longer inter-vehicle distance determination value as the speed of the following vehicle is faster.

Further, the detection crack determination unit may be configured to set the inter-vehicle distance determination value to a preset upper limit value if the speed of the following vehicle exceeds a preset threshold speed.

In addition, if the speed difference between the two vehicles extracted by the determination target vehicle extraction unit exceeds the set speed difference reference value, the detection crack determination unit determines that these two vehicles are not detected crack vehicles. It may be configured to make a determination. If the two vehicles extracted by the determination target vehicle extraction unit are one vehicle that has been erroneously detected due to detection failure, the speeds of these two vehicles are approximately the same. Therefore, if the speed difference between the two vehicles is relatively large (if it is larger than the speed difference reference value), it can be determined that the vehicle is not one that has been erroneously detected due to detection failure (it is two vehicles). As a result, it is possible to improve the accuracy of determining whether the two vehicles extracted by the determination target vehicle extraction section are one vehicle that was incorrectly detected due to detection cracks, and as a result, the accuracy of counting the number of vehicles can be further improved. can be improved.

In addition, if the speed difference between the two vehicles extracted by the determination target vehicle extraction unit exceeds the set speed difference reference value, the detected crack determination unit determines the detected crack using the inter-vehicle distance determination value. It is preferable to have a configuration in which this is not performed. With this configuration, the processing load on the apparatus body will not be increased unnecessarily.

According to the present invention, it is possible to suppress false detection of a vehicle due to cracked detection and improve the accuracy of observing traffic conditions.

Figure 1 (A) is a plan view of the road on which the traffic situation is to be observed, viewed in the vehicle width direction, and Figure 1 (B) is a plan view of the road surface viewed from above the road on which the traffic situation is to be observed. FIG. FIG. 2 is a diagram showing a connection between a traffic situation observation device and a radio radar. FIG. 2 is a block diagram showing the configuration of main parts of the traffic situation observation device. It is a figure showing counting data. It is a figure which shows tracking data. 3 is a flowchart showing the operation of a radio wave radar. 3 is a flowchart showing tracking data generation processing. It is a flowchart which shows counting processing. It is a flowchart which shows the counting process of another example. It is a flowchart which shows the counting process of another example.

Embodiments of this invention will be described below.

<1. Application example＞
FIG. 1 is a schematic diagram showing a driving route for observing traffic conditions with a traffic condition observation device according to this example. Figure 1 (A) is a plan view of the road on which the traffic situation is to be observed, viewed in the vehicle width direction, and Figure 1 (B) is a plan view of the road on which the traffic situation is to be observed, viewed from above. FIG. FIG. 2 is a diagram showing the connection between the traffic situation observation device and the radio radar.

As shown in FIG. 2, a radio wave radar 5 is connected to the traffic situation observation device 1 according to this example. The traffic situation observation device 1 and the radio wave radar 5 may be connected by wire or may be connected wirelessly. Further, as will be described later, the radio wave radar 5 is installed at a position where it can scan a target area 100 set for the travel route with radio waves, which are search waves. That is, the radio wave radar 5 will be installed near the target area 100, but the traffic situation observation device 1 may be installed near the target area 100, or at a control center or data center far from the target area 100. It may also be installed at a collection center or the like.

FIG. 1 shows a three-lane road. The radio wave radar 5 is, for example, a pole installed on the roadside of a running road, or an arm attached to this pole that extends in the width direction (vehicle width direction) of a vehicle 110 (regular vehicle 110a, large vehicle 110b) running on the running road. It is attached to. Moreover, the radio wave radar 5 may be attached to a gantry installed so as to straddle the travel route.

The radio wave radar 5 scans the target area 100 shown in FIG. 1 with radio waves (corresponding to a search wave in the present invention) and detects the position, speed, and size of a traveling vehicle 110. The radio wave radar 5 repeatedly scans the target area 100 using radio waves at set time intervals (for example, at 100 msec intervals).

Furthermore, the radio wave radar 5 may be configured to generate tracking data indicating the travel trajectory of the vehicle 110 within the target area 100. In this example, the radio wave radar 5 will be described as having a configuration for generating this tracking data, but the traffic situation observation device 1 may also be provided with a configuration for generating this tracking data. The configuration for generating this tracking data is such that the vehicle 110 detected in the previous scan of the target area 100 by the radio wave radar 5 is associated with the vehicle 110 detected in the current scan of the target area 100 by radio waves. Every time the radio wave radar 5 scans the target area 100, it outputs vehicle detection data in which the ID, detection time, position, speed, and size of the vehicle 110 detected in the current scan are associated with each other. The detection time may be, for example, the time when the scanning of the target area 100 is started, or the time when the scanning of the target area 100 is finished.

The above-mentioned vehicle detection data is input from the radio wave radar 5 to the traffic situation observation device 1 according to this example.

The traffic situation observation device 1 processes the input vehicle detection data and extracts two vehicles 110 that are consecutively lined up in the traveling direction of the vehicles 110 on the traveling route. The vehicles 110 lined up in the traveling direction are two vehicles 110 detected in the same lane. In other words, the two vehicles 110 detected in different lanes are not vehicles 110 that were lined up in the traveling direction. When the radio radar 5 erroneously detects two vehicles 110 due to detection failure when there is actually one vehicle 110, the two erroneously detected vehicles 110 are lined up in the traveling direction on the traveling route. Furthermore, even if two vehicles 110 are traveling in the same lane, two vehicles 110 that are not lined up consecutively in the traveling direction may be mistakenly detected as two vehicles due to detection errors. isn't it.

The traffic situation observation device 1 extracts combinations of vehicles 110 that may have been falsely detected as two due to detection failure (combinations of vehicles 110 for which it is necessary to determine whether or not they are false detections), and Combinations of vehicles 110 that have no possibility of being erroneously detected as vehicles (combinations of vehicles 110 for which it is unnecessary to determine whether or not they are erroneously detected) are not extracted.

The traffic situation observation device 1 determines whether the two extracted vehicles 110 are one vehicle that was erroneously detected due to a detection error. The traffic situation observation device 1 makes this determination by comparing the extracted inter-vehicle distance between the two vehicles 110 and the inter-vehicle distance determination value. Many drivers drive the vehicle 110 while maintaining a distance between them that allows them to avoid rear-ending the preceding vehicle 110 even if the preceding vehicle 110 suddenly decelerates or stops suddenly. In other words, it is rare for the following vehicle 110 to travel at such a distance that it is impossible to avoid rear-end collision with the preceding vehicle 110. From this, the traffic situation observation device 1 can determine whether one vehicle has been erroneously detected due to detection failure by comparing the inter-vehicle distance between the two extracted vehicles 110 and the inter-vehicle distance judgment value. I can do it.

In this example, the traffic situation observation device 1 sets the inter-vehicle distance determination value according to the speed of the following vehicle 110. As is well known, the higher the speed of vehicle 110, the longer the inter-vehicle distance required to avoid rear-ending collision with preceding vehicle 110. The traffic situation observation device 1 sets the inter-vehicle distance determination value to be longer as the speed of the following vehicle 110 becomes faster (as the speed becomes higher).

Note that the traffic situation observation device 1 may set the inter-vehicle distance determination value according to the average speed of the two vehicles 110, or may set it according to the speed of the preceding vehicle 110. Furthermore, if the speed used to set the inter-vehicle distance determination value exceeds the upper limit speed, the inter-vehicle distance determination value to be set may be the upper limit value. For example, if the speed used to set the inter-vehicle distance determination value is 100 km/h or more, the inter-vehicle distance determination value to be set may be set to an upper limit value, for example, 80 m.

Furthermore, the traffic situation observation device 1 counts the number of vehicles 110 that have traveled in the target area 100 based on the determination result of the detected crack. Therefore, the traffic situation observation device 1 can prevent the accuracy of counting the number of vehicles 110 from decreasing due to detection failure of the vehicles 110. That is, the accuracy of traffic situation observation by the traffic situation observation device 1 can be improved.

<2. Configuration example>
FIG. 3 is a block diagram showing the configuration of the main parts of the traffic situation observation device according to this example. The traffic situation observation device 1 includes a control unit 11, a vehicle detection data input section 12, an observation information storage database 13 (observation information storage DB 13), and an output section 14.

The control unit 11 controls the operation of each part of the traffic condition observation device 1 main body. The control unit 11 also includes a determination target vehicle extraction section 21, a detected crack determination section 22, a counting section 23, and a tracking data generation section 24. Details of the determination target vehicle extraction section 21, detected crack determination section 22, counting section 23, and tracking data generation section 24 included in the control unit 11 will be described later.

Vehicle detection data is input to the vehicle detection data input section 12 from the radio wave radar 5 . The vehicle detection data is data in which the ID, detection time, position, speed, and size of the vehicle 110 are associated with each other. The vehicle detection data input unit 12 may be configured to be connected to the radio wave radar 5 by wire, or may be configured to be connected to the radio wave radar 5 wirelessly. This vehicle detection data input section 12 corresponds to the vehicle detection data acquisition section in the present invention.

The observation information storage DB 13 stores a count data storage section 13a and a tracking data storage section 13b. The observation information storage DB 13 is composed of a storage medium such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or an SD memory card. The count data storage section 13a stores the count data shown in FIG. 4, and the tracking data storage section 13b stores the tracking data shown in FIG.

The observation date of the measurement data shown in FIG. 4 is September 15, 2019. The counting data is data in which the number of vehicles 110 traveling in each lane (right lane, center lane, and left lane) of the target area 100 is registered by vehicle type (regular vehicle, large vehicle) for each time period. Although FIG. 4 shows an example in which the time slots are divided into one hour, the time slots may be of any length, such as 30 minutes or 2 hours. Further, the length of each time period may or may not be constant. For example, for 6:00 to 10:00 and 17:00 to 20:00, the length of the time slot is 30 minutes, for 10:00 to 17:00, the length of the time slot is 1 hour, and for the remaining time (0:00 to 6:00), the length of the time slot is 30 minutes. , hours, and 20:00 to 24:00), the length of the time period may be set to 2 hours.

Further, the types (vehicle models) of the vehicles 110 whose number is counted are not limited to two types, ordinary vehicles and large vehicles, but may be three or more types, or classification by vehicle type may be eliminated.

The observation date of the tracking data shown in FIG. 5 is September 15, 2019. This tracking data is data that associates the ID for identifying the vehicle 110, the size of the vehicle 110, and the speed and position of the vehicle 110 for each detection time. In this example, the position of the vehicle 110 is determined by the distance from the first reference position in the running direction of the vehicle 110 (for example, the distance from the installation position of the radio wave radar 5 in the running direction) and the second reference in the width direction of the running path. The distance from the position (for example, the distance from the center position in the width direction of the running path, where the right side is a positive value and the left side is a negative value). The position in the traveling direction is the position of the head of the vehicle 110. Further, the position in the width direction is the center position of the vehicle 110 in the width direction. The size of the vehicle 110 indicates the overall length (vehicle length) and the overall width (vehicle width).

The output unit 14 outputs counting data, tracking data, etc. as observation information to a higher-level device (not shown) such as a control center.

Next, the determination target vehicle extraction section 21, detected crack determination section 22, counting section 23, and tracking data generation section 24 included in the control unit 11 will be described.

The determination target vehicle extraction unit 21 extracts two vehicles 110 that are consecutively lined up in the traveling direction as the vehicles 110 for which detected cracks are to be determined. The determination target vehicle extraction unit 21 refers to the tracking data (see FIG. 5) of each vehicle 110 obtained by processing the vehicle detection data input from the radio wave radar 5, and identifies the vehicles 110 that are consecutively lined up in the traveling direction. The two vehicles 110 that were present are extracted.

The detected crack determination unit 22 sets an inter-vehicle distance determination value according to the speed of the vehicle 110 following the two vehicles 110 extracted by the determination target vehicle extraction unit 21. For example, the detected crack determination unit 22 sets a value calculated using a function F(v) that uses the speed v of the following vehicle 110 as a variable as the inter-vehicle distance determination value. This function F(v) is set in advance. The function F(v) is any function as long as the value of the calculated inter-vehicle distance judgment value increases as the value of the speed v increases (as the speed of the following vehicle 110 increases). (Not limited to linear or quadratic functions.)

In addition, instead of using the function F(v), a table is stored in which the inter-vehicle distance judgment value is associated with each speed range of the vehicle 110 divided into multiple parts, and this table is used to calculate the inter-vehicle distance judgment value. It is also possible to configure the configuration to be set.

The detected crack determination unit 22 compares the inter-vehicle distance between the two vehicles 110 extracted by the determination target vehicle extraction unit 21 and the inter-vehicle distance determination value set for these two vehicles 110, and determines whether there is a detected crack. Determine whether there is. If the inter-vehicle distance between the two vehicles 110 extracted by the determination target vehicle extraction section 21 is shorter than the inter-vehicle distance determination value set for these two vehicles 110, the detected crack determination unit 22 determines that there is a detected crack. (It is determined that the two vehicles 110 extracted by the determination target vehicle extraction unit 21 are one vehicle.) On the other hand, if the inter-vehicle distance between the two vehicles 110 extracted by the judgment target vehicle extraction section 21 is equal to or greater than the inter-vehicle distance judgment value set for these two vehicles 110, the detection crack determination section 22 detects the crack. It is determined that there is no crack (it is determined that the two vehicles 110 extracted by the determination target vehicle extraction unit 21 are two vehicles).

Furthermore, the detected crack determination unit 22 determines that the two vehicles 110 that have been determined to have detected cracks (the two vehicles 110 extracted by the determination target vehicle extraction unit 21) are large vehicles 110b.

The counting unit 23 counts the number of vehicles 110 based on the determination result of the detected crack determination unit 22, and generates the count data shown in FIG. 4.

The tracking data generation unit 24 processes the vehicle detection data input from the radio wave radar 5 and generates the tracking data shown in FIG. 5.

The control unit 11 of the traffic situation observation device 1 is composed of a hardware CPU, memory, and other electronic circuits. When the hardware CPU executes the traffic situation observation program according to the present invention, it operates as a determination target vehicle extraction section 21, a detected crack determination section 22, a counting section 23, and a tracking data generation section 24. The memory also has an area for expanding the traffic situation observation program according to the present invention and an area for temporarily storing data generated when the traffic situation observation program is executed. The control unit 11 may be an LSI integrated with a hardware CPU, memory, and the like. Further, the hardware CPU is a computer that executes the traffic situation observation method according to the present invention.

Note that since the radio wave radar 5 is well known, a description of its configuration will be omitted.

<3. Operation example>
First, detection of the vehicle 110 by the radio wave radar 5 will be explained. Detection of the vehicle 110 by the radio wave radar 5 is well known, so it will be briefly explained here.

FIG. 6 is a flowchart showing the operation of the radio wave radar. At the scanning timing of the target area 100 (s1), the radio wave radar 5 scans the target area 100 with a radio wave that is an exploration wave, and detects the reflected wave to detect a vehicle 110 located within the target area 100. Detection processing is performed to detect (s2). In s2, the radio wave radar 5 detects the time taken to detect a reflected wave (radio wave flight time) and the frequency of the reflected wave for each radio wave irradiation direction, and detects the object (vehicle 110, road surface, etc.) that reflected the radio wave. ), the size of this object, and the speed of this object.

When the radio radar 5 completes the detection processing in s2, it performs a matching process that associates the vehicle 110 detected in the previous scan of the target area 100 with the vehicle 110 detected in the current scan of the target area 100. (s3). The matching process in s3 is performed using the size, position, and speed of the vehicle 110. The radio wave radar 5 assigns the ID assigned to the vehicle 110 that has been associated with the vehicle 110 that has been associated with the vehicle 110 detected in the previous scan (vehicle 110 that has been detected in the current scan). On the other hand, the radio wave radar 5 generates and assigns a new ID to a vehicle 110 that could not be associated with the vehicle 110 detected in the previous scan (vehicle 110 detected in the current scan).

The radio radar 5 generates vehicle detection data that associates the ID, detection time, position, speed, and size for each vehicle 110 detected in the current scan, and uses the generated vehicle detection data to observe traffic conditions. Output to device 1 (s4) and return to s1.

In this example, the radio wave radar 5 scans the target area 100 with radio waves at intervals of 100 msec, and outputs vehicle detection data regarding the detected vehicle 110 to the traffic situation observation device 1.

Further, by performing the above processing, the radio wave radar 5 can obtain the tracking data shown in FIG. 5 for each vehicle 110 that has traveled in the target area 100.

Furthermore, the traffic situation observation device 1 can generate the tracking data shown in FIG. 5 by classifying and totaling the vehicle detection data input from the radio wave radar 5 by ID.

Further, the radio wave radar 5 may be configured not to generate tracking data. In this case, the radio radar 5 does not perform the process related to s3 described above, but generates vehicle detection data in which position, speed, and size are associated (vehicle detection data without ID association) in s4, and It is output to the traffic condition observation device 1. The traffic situation observation device 1 performs the same process as in s3 described above to generate tracking data.

The traffic situation observation device 1 performs a tracking data generation process for generating tracking data and a counting process for counting the number of vehicles 110 in parallel.

First, the tracking data generation process will be explained. FIG. 7 is a flowchart showing the tracking data generation process.

The traffic situation observation device 1 waits for vehicle detection data to be input from the radio wave radar 5 to the vehicle detection data input unit 12 (s11). In the traffic situation observation device 1, when vehicle detection data is input from the radio radar 5, the tracking data generation unit 24 searches the tracking data stored in the tracking data storage unit 13b using the input ID as a key. , it is determined whether there is tracking data for the corresponding ID (s12). If there is tracking data of the corresponding ID, the tracking data generation unit 24 additionally registers the currently input vehicle detection data to the tracking data (s13), and returns to s11. Furthermore, if there is no tracking data for the corresponding ID, the tracking data generating section 24 generates the tracking data for the corresponding ID, stores the generated tracking data in the tracking data generating section 24 (s14), and returns to s11. The vehicle detection data input this time is registered in the tracking data generated in s14.

The traffic situation observation device 1 generates tracking data for each vehicle 110 that has traveled in the target area 100 by executing the process shown in FIG. 7 for each vehicle detection data input from the radio radar 5, and performs tracking It can be stored in the data storage section 13b.

Note that in the above explanation, the case where vehicle detection data associated with an ID is input from the radio wave radar 5 is taken as an example, but when vehicle detection data with no ID associated is input from the radio wave radar 5. In this step, the tracking data generation unit 24 performs the matching process described in s3 above to generate tracking data.

Next, the counting process will be explained. FIG. 8 is a flowchart showing the counting process.

The determination target vehicle extraction unit 21 extracts two vehicles 110 for which detected cracks are to be determined (s21). In s21, a vehicle 110 that has reached a predetermined passing line in the running direction is extracted as a preceding vehicle 110, and a vehicle 110 that has been lined up continuously in the running direction with respect to this preceding vehicle 110 is extracted as a following vehicle 110. . The preceding vehicle 110 and the following vehicle 110 are vehicles 110 traveling in the same lane. Further, there is no other vehicle 110 that was traveling in the same lane between the preceding vehicle 110 and the following vehicle. The passing line is a line that defines the position of the vehicle 110 in the traveling direction, and is a line that extends in the vehicle width direction. The passing line is set within the target area 100. Further, the passage line is set on the exit side of the target area 100.

The determination target vehicle extraction unit 21 refers to the tracking data stored in the tracking data storage unit 13b and extracts two vehicles 110 for which detection cracks are to be determined. As described above, the tracking data is registered vehicle detection data of the vehicle 110 detected by the radio wave radar 5. That is, the determination target vehicle extraction unit 21 processes the vehicle detection data of the vehicles 110 detected by the radio wave radar 5 and extracts two vehicles 110 for which detection cracks are to be determined.

The detected crack determination unit 22 detects the currently extracted speed v of the following vehicle 110 (s22). The detected crack determination unit 22 detects the speed v of the following vehicle 110 by referring to the tracking data of the following vehicle 110. Further, the speed v of the following vehicle 110 may be the speed v detected for the following vehicle 110 when the preceding vehicle 110 reaches the passing line, or the speed v detected for the following vehicle 110 by that time. It may be determined by the average value of the calculated speeds, or by other methods.

The detected crack determination unit 22 calculates and sets an inter-vehicle distance determination value using a function F(v) that uses the speed v of the following vehicle 110 detected in s22 as a variable (s23).

The detected crack determination unit 22 calculates the inter-vehicle distance between the currently extracted preceding vehicle 110 and the following vehicle 110 (s24). In s24, vehicle detection data of the preceding vehicle 110 and vehicle detection data of the following vehicle 110, which have the same detection time, are used. In s24, the length obtained by subtracting the vehicle length of the preceding vehicle 110 from the distance between the position of the preceding vehicle 110 in the running direction and the position of the following vehicle 110 in the running direction, which have the same detection time, is calculated. It is calculated as the inter-vehicle distance to the vehicle 110. That is, the detected crack determination unit 22 calculates the inter-vehicle distance between the preceding vehicle 110 and the following vehicle 110 by subtracting the vehicle length of the preceding vehicle 110 from the inter-vehicle distance between the preceding vehicle 110 and the following vehicle 110.

Note that the inter-vehicle distance between the preceding vehicle 110 and the following vehicle 110 may be calculated using a plurality of sets of vehicle detection data of the preceding vehicle 110 and vehicle detection data of the following vehicle 110 that have the same detection time. . In this case, the inter-vehicle distance between the preceding vehicle 110 and the following vehicle 110 is calculated for each pair having the same detection time, and the average value, minimum value, etc. of the inter-vehicle distance is calculated between the preceding vehicle 110 and the following vehicle to be used in the following process. The distance between the vehicle 110 and the vehicle 110 may be used.

The detected crack determination unit 22 compares the inter-vehicle distance determination value set in s23 with the inter-vehicle distance between the preceding vehicle 110 and the following vehicle 110 calculated in s24 (s25). If the inter-vehicle distance between the preceding vehicle 110 and the following vehicle 110 is shorter than the inter-vehicle distance determination value, the detected crack determination unit 22 determines that there is a detected crack (s26), and distinguishes between the preceding vehicle 110 and the following vehicle 110 extracted in s21. It is counted as one vehicle 110 (s27). On the other hand, if the inter-vehicle distance between the preceding vehicle 110 and the following vehicle 110 is equal to or greater than the inter-vehicle distance determination value, the detected crack determination unit 22 determines that there is no detected crack (s29), and selects one preceding vehicle 110 extracted in s21. It is counted as the vehicle 110 (s30), and the process returns to s21. In s30, it is not determined whether or not the following vehicle 110 extracted in s21 has a detected crack. Regarding the following vehicle 110, when the vehicle 110 is extracted as the preceding vehicle 110 in s21, it is determined whether or not there is a detection crack.

Further, if the tracking data generation unit 24 determines that there is a detection crack in s26, it corrects the tracking data of the corresponding vehicle 110 (s28) and returns to s21. In s28, the size of the tracking data of the preceding vehicle 110 is corrected. Specifically, the vehicle length of the preceding vehicle 110 is corrected to a value obtained by adding the vehicle length of the following vehicle 110. Furthermore, it is registered in the tracking data of the following vehicle 110 that it is a false detection due to a crack in the detection.

Note that the tracking data of the following vehicle 110 may be deleted when the following vehicle 110 is no longer detected.

In this way, the traffic situation observation device 1 according to this example can prevent the accuracy of counting the number of vehicles 110 from decreasing due to detection failure of the vehicles 110. In other words, the traffic situation observation device 1 can improve the accuracy of counting the number of vehicles 110 that have traveled in the target area 100. In other words, the accuracy of traffic situation observation by the traffic situation observation device 1 can be improved.

The counting process described above may be executed in real time in response to the input of vehicle detection data from the radio wave radar 5, or may be executed as a batch process at an appropriate timing.

Furthermore, when the road is congested, it is preferable to count the number of vehicles 110 without determining the above-described detection failure. The presence or absence of traffic congestion may be determined based on the average speed of vehicles 110 traveling in target area 100. For example, when the average speed is less than the traffic jam determination speed (for example, 5 km/h or 10 km/h), it may be determined that a traffic jam is occurring.

<4. Modified example>
・Modification example 1
The counting process of the traffic situation observation device 1 according to the example described above may be replaced with the process shown in FIG. In FIG. 9, the same steps as those in FIG. 8 are given the same step numbers. In this first modification, in s21, the determination target vehicle extraction unit 21 extracts two vehicles 110 for which detected cracks are to be determined. The detection crack determination unit 22 detects the speeds of the two vehicles 110 (the preceding vehicle 110 and the following vehicle 110) extracted in s21 (s41), and calculates the absolute value of the speed difference between these two vehicles 110. (s42). The detected crack determination unit 22 determines whether the speed difference (absolute value) calculated in s42 is less than the set speed difference reference value (s43).

If the two vehicles 110 extracted in s21 are traveling at a certain speed difference, these two vehicles 110 are different vehicles. If the detected crack determination unit 22 determines that the speed difference calculated in s42 is not less than the set speed difference reference value, it determines that the currently extracted preceding vehicle 110 does not have a detected crack (s29), and performs the processing in s30. and return to s21.

Further, if the detected crack determination unit 22 determines that the speed difference (absolute value) calculated in s42 is less than the set speed difference reference value, it performs the processes in s23 to s30 described above and returns to s21. .

In this way, the traffic situation observation device 1 according to this modification 1 uses the speed difference to determine in advance whether or not there is a detection crack for the two vehicles 110 extracted in s21. If it is not determined that there is no detected crack, the above-mentioned processes from s23 to s30 are performed. That is, when the traffic situation observation device 1 according to the first modification determines that there is no detection crack using the speed difference between the two vehicles 110, it does not perform the processing related to s23 to s30 described above. Therefore, the processing load on the main body of the traffic condition observation device 1 can be suppressed. Furthermore, the traffic situation observation device 1 according to the first modification can improve the accuracy of counting the number of vehicles 110 that have traveled through the target area 100, similarly to the above example.

・Modification 2
Further, the counting process shown in FIG. 8 or 9 may be replaced with the process shown in FIG. 10. In FIG. 10, the same step numbers are assigned to the same processes as those shown in FIGS. 8 and 9.

In this second modification, in s21, the determination target vehicle extraction unit 21 extracts two vehicles 110 for which detected cracks are to be determined. The detected crack determination unit 22 determines whether the two vehicles 110 extracted in s21 are both large vehicles (s51). When a large vehicle is erroneously detected as two vehicles 110 due to detection failure, at least one of the vehicles 110 is detected as an ordinary vehicle rather than a large vehicle. That is, if the two vehicles 110 extracted in s21 are both large vehicles, it can be determined that no detected crack has occurred.

When the detected crack determination unit 22 determines that the two vehicles 110 extracted this time are both large vehicles, it determines that the preceding vehicle 110 extracted this time is not detected cracked (s29), and performs the processing related to s30. , return to s21.

Furthermore, if at least one of the two vehicles 110 extracted in s21 is not a large vehicle, the detected crack determination unit 22 performs the processes in s23 to s30 described above, and returns to s21.

In this way, the traffic situation observation device 1 according to the second modification performs a preliminary judgment on the two vehicles 110 extracted in s21 as to whether or not there is a detection crack based on the size of the vehicle 110, and If it is not determined that there is no detected crack, the above-mentioned processes from s23 to s30 are performed. Therefore, the processing load on the main body of the traffic condition observation device 1 can be suppressed. Furthermore, the traffic situation observation device 1 according to the first modification can improve the accuracy of counting the number of vehicles 110 that have traveled through the target area 100, similarly to the above example.

Furthermore, the traffic situation observation device 1 may have a configuration in which it is not connected to the radio wave radar 5. For example, detection data of the vehicle 110 detected by the radio wave radar 5 or the like is stored in a storage medium such as an HDD or an SSD. This storage medium may be connected to the traffic situation observation device 1, and the detection data of the vehicle 110 may be acquired from the storage medium to which the traffic situation observation device 1 is connected.

It should be noted that the present invention is not limited to the above-described embodiments as they are, but can be implemented by modifying the constituent elements within the scope of the invention at the implementation stage. Moreover, various inventions can be formed by appropriately combining the plurality of components disclosed in the above embodiments. For example, some components may be deleted from all the components shown in the embodiments. Furthermore, components from different embodiments may be combined as appropriate.

Furthermore, the correspondence between the configuration according to the present invention and the configuration according to the above-described embodiments can be described as in the following supplementary notes.
<Additional notes>
A vehicle detection data acquisition unit that acquires vehicle detection data including the position and speed of the vehicle (110) detected by scanning the target area (100) set on the travel path of the vehicle (110) with a search wave. (12) and
a determination target vehicle extraction unit (21) that processes the vehicle detection data acquired by the vehicle detection data acquisition unit (12) and extracts two vehicles (110) that were consecutively lined up in the travel direction on the travel path; )and,
Detection crack detection for determining whether or not these two vehicles (110) are detection crack vehicles using the speeds and inter-vehicle distances of the two vehicles (110) extracted by the determination target vehicle extraction unit (21). A determination unit (22);
The detection crack determination unit (22) sets an inter-vehicle distance determination value according to the speed of the vehicle extracted by the determination target vehicle extraction unit (21), and sets an inter-vehicle distance determination value according to the speed of the vehicle extracted by the determination target vehicle extraction unit (21). If the inter-vehicle distance between the two vehicles (110) is shorter than the inter-vehicle distance determination value, the traffic situation observation device (1) determines that these two vehicles (110) are one vehicle (110) that has failed to detect the two vehicles (110). ).

1...Traffic situation observation device 5...Radio wave radar 11...Control unit 12...Vehicle detection data input unit 13...Observation information storage database (observation information storage DB)
13a... Counting data storage section 13b... Tracking data storage section 14... Output section 21... Judgment target vehicle extraction section 22... Judgment section 23... Counting section 24... Tracking data generation section 100... Target area 110... Vehicle 110a... Ordinary vehicle 110b... large vehicle

Claims (6)

a vehicle detection data acquisition unit that acquires vehicle detection data including a vehicle position and speed detected by scanning a target area set on a vehicle travel route with a search wave;
a determination target vehicle extraction unit that processes the vehicle detection data acquired by the vehicle detection data acquisition unit and extracts two vehicles that were consecutively lined up in the travel direction on the travel path;
If the inter-vehicle distance between the two vehicles extracted by the determination target vehicle extraction unit is shorter than the set inter-vehicle distance determination value, the detection crack determination determines that these two vehicles are one vehicle with the detection crack. and,
The detection crack determination unit sets the inter-vehicle distance determination value to be longer as the speed of the vehicle following the two vehicles extracted by the determination target vehicle extraction unit is faster, and sets a threshold value to which the speed of the following vehicle is set. A traffic condition observation device that sets the inter-vehicle distance judgment value to an upper limit value if the speed exceeds the speed limit . The detection crack determination unit determines that these two vehicles are not detection crack vehicles if the speed difference between the two vehicles extracted by the determination target vehicle extraction unit exceeds a set speed difference reference value. The traffic situation observation device according to claim 1 , which makes a determination. If the speed difference between the two vehicles extracted by the determination target vehicle extraction unit exceeds a set speed difference reference value, the detected crack determination unit determines the detected crack using the inter-vehicle distance determination value. The traffic situation observation device according to claim 2 , which does not make a determination. The traffic situation observation device according to any one of claims 1 to 3 , further comprising a counting unit that counts the number of vehicles that have traveled within the target area based on the determination result of the detection crack determination unit. The vehicle detection data, including the position and speed of the vehicle detected by scanning the target area set on the vehicle travel route with a search wave, is processed, and the vehicle detection data including the vehicle position and speed is processed to determine whether the vehicle is continuously lined up in the travel direction on the travel route. a determination target vehicle extraction step of extracting two vehicles;
If the inter-vehicle distance between the two vehicles extracted in the determination target vehicle extraction step is shorter than the set inter-vehicle distance determination value, the detection crack determination determines that these two vehicles are one vehicle with the detection crack. The computer executes the steps,
In the detection crack determination step, the faster the speed of the vehicle following the two vehicles extracted in the determination target vehicle extraction step, the longer the inter-vehicle distance determination value is set, and the threshold value to which the speed of the following vehicle is set is set. If the vehicle speed exceeds the speed limit, the traffic condition observation method includes the step of setting the inter-vehicle distance determination value to an upper limit value . The vehicle detection data, including the position and speed of the vehicle detected by scanning the target area set on the vehicle travel route with a search wave, is processed, and the vehicle detection data including the vehicle position and speed is processed to determine whether the vehicle is continuously lined up in the travel direction on the travel route. a determination target vehicle extraction step of extracting two vehicles;
If the inter-vehicle distance between the two vehicles extracted in the determination target vehicle extraction step is shorter than the set inter-vehicle distance determination value, the detection crack determination determines that these two vehicles are one vehicle with the detection crack. make the computer perform the steps and,
In the detection crack determination step, the faster the speed of the vehicle following the two vehicles extracted in the determination target vehicle extraction step, the longer the inter-vehicle distance determination value is set, and the threshold value to which the speed of the following vehicle is set is set. The traffic situation observation program is a step of setting the inter-vehicle distance judgment value to an upper limit value if the speed exceeds the speed .

Images