JP2014099123A - Vehicle detection device and vehicle detection method - Google Patents

Abstract

A shape characteristic of a vehicle to be determined is determined.
A sensor unit that is installed above a road and detects a road and a vehicle to be determined as a detection signal by transmitting and receiving radio waves at predetermined time intervals, and a sensor unit that is vertically above the sensor unit and on a surface of the road. A first distance between the first position and the second position where the radio wave that has passed close to the rear end of the traveling vehicle in the upper direction and the traveling direction is reflected on the road; and from the first position to the second position A second distance to the front end in the traveling direction of the vehicle to be judged is acquired, and the calculation is performed using the first distance and the second distance acquired at two different times. A vehicle shape feature determination unit that obtains a detection length related to a length of a determination vehicle and a detection height related to a height of the determination target vehicle, and determines a shape characteristic of the determination target vehicle based on the detection length and the detection height; Have.
[Selection] Figure 4

Description

  The present embodiment relates to a vehicle detection device and a vehicle detection method.

    In recent years, research and development of the Intelligent Transport System (ITS) relating to road traffic has been promoted.

  In this field, it is known that a moving direction such as a right turn, a left turn, or a straight line at a vehicle intersection is detected by a video camera. Thereby, at the intersection, when traveling in a direction other than the movement direction permitted in the traveling lane of the vehicle is detected, it can be recognized as a traffic violation act.

  In addition, it is known that a fan beam of a millimeter wave radar installed above a road is used to track the vehicle for every scan in the direction of the azimuth angle of the fan beam. Thus, information can be provided so that the driver can recognize the position and speed of other vehicles and the like even when the road prospect is poor.

  It is also known to detect the position and height of a vehicle by emitting radio from a radio transmission / reception unit installed above one lane of the road and receiving radio reflected by a vehicle. . In addition to this, it is known that the vehicle speed is measured using a Doppler sensor, the vehicle length is calculated from the vehicle speed, and the vehicle type can be roughly determined from the vehicle height and the vehicle length as either a large vehicle or a small vehicle.

  As one structure of the millimeter wave radar, it is known that the distance can be calculated from the delay time of the reception wave with respect to the transmission wave, and the velocity can be measured from the Doppler shift. (For example, refer to Patent Documents 1 to 3 and Non-Patent Document 1)

JP 2000-90389 A JP 2002-99986 A JP 2011-257814 A

Toshiki Yamawaki and Shinichi Yamano, “Millimeter Wave Radar for 60 GHz Automotive”, Fujitsu Ten Technical Report, Vol. 15, no. 2, pages 9 to 18, 1997

  In the technology described in the background art, it is necessary to construct a complicated system in order to measure the vehicle speed using the Doppler sensor and calculate the vehicle length from the vehicle speed, which increases the difficulty and cost.

  With the technology described in the background art, even if a large vehicle or a small vehicle can be roughly determined as a vehicle type from the obtained vehicle height and vehicle length, it is not possible to detect the vehicle type and the vehicle type based on the vehicle shape.

  In one aspect, an object of the present invention is to determine a shape characteristic of a vehicle to be determined.

  According to one aspect of the disclosed vehicle detection device, a sensor unit that is installed above a road and detects a road and a vehicle to be determined as a detection signal by transmitting and receiving radio waves every predetermined time; The first distance between the first position on the surface of the road and the second position where the radio wave that has passed close to the rear end of the traveling vehicle in the upper part and the traveling direction is reflected on the road, And the 2nd distance from the 1st position to the front end of the advancing direction of the to-be-determined vehicle is acquired, and the 1st distance and the 2nd distance acquired for every two different said predetermined time are acquired. And calculating a detection length related to the length of the vehicle to be determined and a detection height related to the height of the vehicle to be determined, and determining a feature of the shape of the vehicle to be determined based on the detection length and the detection height. A vehicle shape feature determination unit. Apparatus is provided.

  According to the disclosed technology, it is possible to determine the characteristics of the shape of the vehicle to be determined.

The schematic plan view which looked at the installation example of the vehicle detection apparatus by an Example from the upper direction of the intersection. The schematic side view which looked at the example of installation of the vehicle detection apparatus shown in FIG. 1 from the side of the intersection. An example of oncoming vehicle information provided through a right turn monitor. The block diagram which shows the structural example of a vehicle detection apparatus. Block diagram showing a configuration example of a sensor device An example of a fan beam emitted from a sensor unit when the fan beam is irradiated while scanning the sensor unit. An example of a fan beam emitted from a sensor unit when a fixed sensor unit that does not scan is installed for each lane of travel. The flowchart of the level determination in a vehicle detection apparatus. An example of the graph which shows a road surface reflected signal level and an apparatus noise level. An example of the detection signal drawn on the coordinate which consists of the vertical axis | shaft of a position, and the horizontal axis of time. Another example of the detection signal drawn on the coordinate consisting of the vertical axis of position and the horizontal axis of time. An example of the graph of the detection signal which detected the vehicle currently drive | working within the detection range within predetermined time. An example of a detection signal obtained in a detection range within a predetermined time. The flowchart of the calculation of a vehicle length and a vehicle height, determination of the shape of a vehicle, and determination of a vehicle type. Explanatory drawing explaining the method to calculate detection length and detection length. The other example of the detection signal obtained in the detection range within the predetermined time. The graph which represented the change of the detection length and detection height which were calculated for every adjacent predetermined time corresponding to the position on a road.

  Hereinafter, embodiments will be described in detail with reference to the drawings. In all the drawings for explaining the embodiments, the same reference numerals are used for those having the same function, and repeated explanation is omitted.

[Example]
<Installation position of vehicle detection device>
In the ITS field, a driving safety support system (Driving Safety Support System (DSSS)) is a vehicle-to-vehicle communication performed between a plurality of vehicles, or a road-to-vehicle communication performed between a certain vehicle and a road-side system. By trying to prevent traffic accidents. The vehicle detection device in the embodiment can be used as an example of DSSS.

  FIG. 1 is a schematic plan view of an installation example of a vehicle detection device according to an embodiment as viewed from above an intersection. With reference to FIG. 1, the installation position on the road 15 of the vehicle detection apparatus 1 in an Example is demonstrated.

  The vehicle detection device 1 is used, for example, in order to avoid a collision accident between a right turn car 73 at an intersection 13 and a counter car 77 that goes straight ahead and faces this. Collision accidents between the right turn vehicle 73 and the oncoming vehicle 77 are statistically frequently occurring. In addition, since the vehicles moving forward collide, the collision energy is high and the possibility of a serious accident is statistically high. Therefore, it is desired to prevent a collision accident between the right turn vehicle 73 and the oncoming vehicle 77.

  In order to prevent accidents at intersections, the inventors have realized that a technique for tracking the position of an oncoming vehicle and a technique for determining the type of vehicle on the opposite lane are required. Furthermore, as a car model, it is not a rough car type such as a large car or a small car, for example, in the case of a rectangular parallelepiped shape such as a bus and in the case where the loading platform is missing from a rectangular parallelepiped like an empty truck, I noticed that the outlook from the right turn car was different. For example, if the vehicle has a shape lacking from a rectangular parallelepiped as in the case where the truck bed at the head of the oncoming lane is empty, there is a possibility that the straight turn car 77 can be seen from the right turn car 73. Further, if the vehicle at the head of the oncoming lane is a rectangular parallelepiped vehicle such as a bus, the possibility that the straight-turned vehicle 77 can be seen from the right turn vehicle 73 can be reduced compared to the former.

  The vehicle detection device 1 in the embodiment includes a sensor device 2 and a signal processing device 3. The sensor device 2 includes a sensor unit 21 and a sensor control unit 23. The sensor unit 21 is installed, for example, on the front side of the right turn vehicle 73 with respect to the center of the intersection 13 on the extension of the traveling lane of the oncoming vehicle 77 and at a high position, and the detection part of the sensor unit 21 is the oncoming vehicle. 77 and the intersection 13 are installed so that they can be seen from the front. In other words, the sensor unit 21 is installed above the opposite side road 157 of the right turn car 73. The installation position P0 in the extending direction of the opposite side road 157 of the sensor unit 21 may be, for example, about 30 m from the center position P1 of the intersection 13 toward the traveling direction of the oncoming vehicle 77. By disposing the sensor unit 21 in this way, the detection range DA from approximately the center position P1 of the intersection 13 to the position P4 in the direction of the oncoming vehicle 77 on the opposite roadway 157 can be installed. For example, the position P1 to the position P4 may be about 120m.

  As described above, by installing the sensor unit 21, it is possible to look down on a range where the oncoming vehicle 77 in which the right turn vehicle 73 may collide exists, and it is possible to detect while tracking the moving position of the oncoming vehicle 77.

  For example, the sensor unit 21 may be installed on the support 16 that can hold the sensor unit 21 above the road 15. Alternatively, if there is an existing standing object on which a road sign is posted, it may be attached and installed. In the sensor device 2, the installation position of the sensor control unit 23 may be a position where input / output with the sensor unit 21 can be performed. The installation position of the sensor control unit 23 may be on the support column 16 or any position of the sidewalk 17, the central separation band 18, the roadside band 19, and the like.

  The installation position of the signal processing device 3 may be a position where the output from the sensor device 2 having the sensor unit 21 and the sensor control unit 23 can reach. The installation position of the signal processing device 3 may be on the support column 16 or any position of the sidewalk 17, the central separation band 18, the roadside band 19, and the like.

<Information Provided by Vehicle Detection Device and Vehicle Detection Method>
FIG. 2 is a schematic side view of the installation example of the vehicle detection device 1 shown in FIG. 1 as viewed from the side of the intersection 13. In FIG. 2, the vehicle detection device 1 detects the positions, vehicle heights, vehicle lengths, and vehicle speeds of the oncoming vehicle 77 and the rearward oncoming vehicle 78 as will be described later. The shape and vehicle speed of the vehicle can be detected. Information on these oncoming vehicles 77 and the like is sent to the communication center 5, for example. The communication center 5 sends information such as the oncoming vehicle 77 to the information providing device 6 installed in the vicinity of the road 15. The information providing device 6 provides information on the oncoming vehicle 77 to the right turn car 73 etc. by wireless etc., so that the occupant of the right turn car 73 passes through a monitor or the like provided on the right turn car 73 and the shape, vehicle type, Information such as vehicle height, vehicle length, and vehicle speed can be obtained.

  FIG. 3 is an example of information on the oncoming vehicle 77 and the like provided to a monitor or the like of the right turn vehicle 73. In the screen divided into four, the screen on the upper right shows the current intersection position by the car navigation system or the like. In the upper left screen, an image of an optical camera (not shown) attached to the sensor unit 21 is shown. The lower right screen shows information on the oncoming vehicle 77 and the like provided from the vehicle detection device 1, and the lower left screen shows an information processing device (not shown) mounted on the right turn car 73 based on the information on the oncoming vehicle 77 and the like. Shows a message that the start of a right turn determined by is dangerous. On the lower right screen, as information on the oncoming vehicle 77 and the like provided from the vehicle detection device 1, for example, a distance from the intersection center P1 to the oncoming vehicle 77, a traveling speed, a vehicle type, and the like are provided. In FIG. 3, information on three oncoming vehicles 77 and the like is provided, and the presence of the B vehicle, which is a small passenger car that is difficult to see behind the A vehicle, which is a large vehicle truck, is further behind. The existence of the C vehicle, which is a passenger car of a small car trying to reach the intersection 13 at a high speed of 70 km / h, is informed.

  As described above, the vehicle detection device and the vehicle detection method according to the embodiment can be used even when there are a plurality of oncoming vehicles. When there is a rear oncoming vehicle 78 traveling behind the oncoming vehicle 77, the rear oncoming vehicle 78 is behind the oncoming vehicle 77, and its presence and shape may not be visually determined from the right turn car 73 side. Such information of the backward oncoming vehicle 78 can also be detected. In this way, the passenger of the right turn vehicle 73 can obtain information such as the oncoming vehicle 77 and the rear oncoming vehicle 78. As a result, the right turn vehicle 73 can safely turn right while avoiding a collision with the oncoming vehicle 77 and the rear oncoming vehicle 78.

  The example in which the information provided by the vehicle detection device and the vehicle detection method according to the embodiment is used by the right turn car 73 has been described above. However, the information provided by the vehicle detection device and the vehicle detection method is, for example, that of the oncoming vehicle 77. It can also be used to inform the traffic situation at the intersection 13 to a vehicle or a traffic control center traveling behind. In addition, in order to inform the bicycle or pedestrian waiting for a signal of the approach of the vehicle to the intersection 13, the information provided by the vehicle detection device 1 can be provided in the form of, for example, voice. As described above, the vehicle detection device and the vehicle detection method according to the embodiments can be used for traffic safety or other purposes.

<Configuration of vehicle detection device 1>
FIG. 4 is a block diagram illustrating a configuration example of the vehicle detection device. With reference to FIG. 4, the vehicle detection device 1 includes a sensor device 2 and a signal processing device 3. FIG. 5 shows a configuration example of the sensor device 2.

<Configuration of sensor device 2>
With reference to FIG. 4, the sensor device 2 acquires information related to an object on the road, and includes a sensor unit 21 and a sensor control unit 23. For example, the sensor unit 21 transmits radio waves toward the road and receives radio waves reflected by the road 15 and the vehicles 7 or objects on the road. The radio wave used in the sensor unit 21 is, for example, a millimeter wave. When millimeter waves are used, it is said that they have stable detection performance regardless of day and night or bad weather. Its performance is as follows: (1) It is possible to detect vehicles with a distance of 100m or more even when rain, snow, fog, or snowstorm (when using a mid-range radar), and (2) the applicable range of environment is rain with a rainfall of 50mm / h or more. It is said that it can cope with visibility of several tens of meters or less during fog or snowstorm, and (3) the allowable range of vibration is ± 8 cm or less (± 1.5 ° or less) in terms of displacement. Therefore, it is suitable for an environment where there is a lot of vibration due to road traffic and it is exposed to wind and rain.

  The sensor unit 21 is, for example, a Frequency Modulated Continuous Wave (FM-CW) type radar. Since transmission is continuously performed while periodically changing the frequency of the radio wave to be transmitted, radio waves having different frequencies are transmitted when a reception wave reflected by an object is received. The distance can be known using the frequency difference between these transmissions and receptions. This FM-CW radar can determine the position of an object in the plane or space. Further, the sensor unit 21 may determine the speed of the moving vehicle 7 using the difference between the upward and downward frequency changes.

  With reference to FIG. 5, the sensor device 2 includes a sensor unit 21 and a sensor control unit 23. The sensor unit 21 includes an antenna 211, a transmission / reception selector switch 213, an antenna connection switch 214, a reception side amplifier 215, a reception side A / D converter 216, a transmission side amplifier 218, and a transmission side D / A converter 219. Have The antenna 211 transmits radio waves, and receives various radio waves that have been reflected back to the road 15 or the vehicle 7 to detect various data regarding the road 15 or the vehicle 7. When the sensor device 2 is scanned (see FIG. 6), it further includes a scanning motor 212. The scanning can be realized by a method using a control device (not shown) for controlling the scanning motor 212 and a power source (not shown). The control device and the power source may be in the sensor control unit 23 or the like. Note that electronic scanning may be performed without using the scanning motor -212.

  The sensor control unit 23 has a control function for communicating the output signal of the sensor unit 21 with the outside. When the sensor control unit 23 operates the transmission / reception changeover switch 213 to switch to transmit radio waves, the digital signal output from the sensor control unit 23 based on the control signal input from the signal processing device 3 to the sensor control unit 23. The signal is sent to the transmission side D / A converter 219. This digital signal is converted into an analog signal by the transmission-side D / A converter 219. The converted analog signal is sent to the transmission side amplifier 218 and amplified. Here, the amplified analog signal may be sent to the receiving side A / D converter 216. This is because the difference between the upward and downward frequency changes is used to determine the speed of the moving vehicle, etc., in order to acquire the device noise level described later, and so on. The amplified analog signal is output as a radio wave from the antenna 211 via the transmission / reception selector switch 213.

  On the other hand, when the sensor control unit 23 operates the transmission / reception changeover switch 213 to switch to receive radio waves, the radio waves received by the antenna 211 pass through the transmission / reception changeover switch 213 and the antenna connection switch 214 as signals. Then, it is inputted to the receiving side amplifier 215 and amplified. The amplified signal is converted into a digital signal by the reception side A / D converter 216 and input to the sensor control unit 23. The sensor control unit 23 outputs a detection signal DS including this digital signal to the signal processing device 3.

  In addition, the sensor control unit 23 can disconnect the connection between the transmission / reception selector switch 213 and the reception-side amplifier 215 by operating the antenna connection switch 214. Note that the sensor unit 21 and the sensor control unit 23 may be separate or integrated.

  The radio wave emitted by the sensor unit 21 is, for example, the fan beam FB. The outer shape of the irradiation range on the road 15 of the fan beam FB has a substantially fan shape.

  FIG. 6 illustrates an example of the fan beam FB irradiated from the sensor unit 21 when the fan beam FB is irradiated while the sensor unit 21 is scanned. In this way, when the road 15 is so wide that one fan beam FB cannot radiate radio waves to all the one side roads of the road, the sensor unit 21 is periodically moved in the width direction of the road, that is, in the azimuth angle direction. Can be scanned. Furthermore, when the range to be detected in the road extending direction cannot be irradiated with one fan beam FB, scanning may be performed in the road extending direction.

  When scanning the fan beam FB, for example, if the scanning cycle is 50 ms, the vehicle 7 traveling at a speed of 200 km / h can be detected. As described above, by using only one sensor unit 21 using scanning, a mechanism for integrating the plurality of sensor units 21 can be omitted, and the cost can be reduced. In this case, the scanning motor may be mounted on the sensor unit 21, and the control device and power source of the scanning motor may be mounted on the sensor control unit 23.

  FIG. 7 illustrates an example of the fan beam FB emitted from the sensor unit 21 when a fixed sensor unit 21 that is not scanned is installed for each lane. In this way, one fixed sensor unit 21 that does not scan may be installed corresponding to one traveling lane. In this case, although the motor for scanning etc. can be omitted, for example, the sensor units 21 for the number of traveling lanes are prepared. The sensor unit 21 is suitable for installing the sensor unit 21 when the speed of the vehicle 7 to be detected is very fast, or in an environment in which bad weather is normal. The sensor control unit 23 has a control function for communicating the output signal of the sensor unit 21 with the outside.

<Configuration of Signal Processing Device 3>
Returning to FIG. 4, the signal processing device 3 includes a signal detection unit 31 and a tracking unit 32. The signal detection unit 31 includes a signal input control unit 311, a level determination unit 312, a reference level storage unit 313, a determination information storage unit 314, a communication unit 318, and a vehicle shape feature determination unit 319. You may have the number recognition part 33. FIG. Here, the vehicle shape feature determination unit 319 may include a vehicle length vehicle height calculation unit 315, a vehicle length vehicle height storage unit 316, and a vehicle shape vehicle type determination unit 317.

  The signal processing device 3 transmits and receives signals to and from the sensor device 2. As will be described later, the signal input control unit 311 outputs control signals for performing various controls on the sensor control unit 23 of the sensor device 2. At the same time, the signal input control unit 311 outputs an apparatus noise level NL acquisition or update signal to the reference level storage unit 313 as will be described later. An output signal from the sensor control unit 23 is input to the level determination unit 312 of the signal detection unit 31. The signal input to the level determination unit 312 is input to the reference level storage unit 313 and simultaneously stored in the determination information storage unit 314. Of the signal data stored in the determination information storage unit 314, data specified by the tracking unit 32 described later is read into the vehicle length vehicle height calculation unit 315 of the vehicle shape feature determination unit 319. The vehicle length vehicle height calculation unit 315 calculates the read data, calculates the vehicle height, vehicle length, and vehicle speed, and outputs the result to the vehicle length vehicle height storage unit 316 of the vehicle shape feature determination unit 319.

  In the vehicle shape feature determination unit 319, the vehicle shape vehicle type determination unit 317 reads data stored in the vehicle length vehicle height storage unit 316, and determines the vehicle shape and vehicle type. The communication unit 318 converts the result data calculated by the vehicle length vehicle height calculation unit 315 and the result data determined by the vehicle shape vehicle type determination unit 317 of the vehicle shape feature determination unit 319 into a signal that can be received externally. It transmits to the communication center 5 which is not illustrated in FIG.

  Note that the vehicle number recognition unit 33 may be included in the signal processing device 3. The vehicle number recognition unit 33 uses a known optical character reader (OCR) to display the size of the automobile registration number mark or vehicle number mark (hereinafter referred to as a license plate) and the classification number among the numbers written on the number plate. May be used to make the content of the result data determined by the vehicle shape vehicle type determination unit 317 highly accurate. Since the first digit of the classification number indicates the following vehicle type, it can be used to make the details of the vehicle type determination highly accurate or enable more detailed vehicle type determination. For example, in the case of an automobile registration number mark, the first digit numbers 1, 2, 3, 4, 5, 6, 7, 8, 9, 0 are respectively ordinary freight cars, ordinary passenger cars, ordinary passenger cars, and small freight cars. , Small passenger cars, small freight cars, small passenger cars, special purpose automobiles, large special automobiles, and construction machinery. In place of the vehicle number recognition unit 33, an external vehicle number recognition device (not shown) and the signal detection unit 31 are connected so that information on the vehicle number of the oncoming vehicle 77 is input to the signal detection unit 31. You may comprise.

  The tracking unit 32 specifies the data of the vehicle 7 to be subjected to shape calculation and vehicle type determination based on the determination data obtained by the level determination unit 312 and the like, and stores the data in the determination information storage unit 314. As the signal processing device 3, a computer including a processor, a memory and the like can be used. Further, a known tracking device can be used as the tracking unit 32. As the vehicle number recognition unit 33, a known vehicle number recognition system or the like can be used.

<Operation of the vehicle detection device 1>
FIG. 8 is a flowchart of level determination in the vehicle detection device 1. Below, the operation | movement which implement | achieves a vehicle detection method with the vehicle detection apparatus 1 is demonstrated. FIG. 8 illustrates a case where the fan beam FB is irradiated while the sensor unit 21 is scanned (see FIG. 6). When a fixed sensor unit 21 that does not scan is installed for each lane (see FIG. 7), the sensor unit 21 is scanned except that steps S24 and S25 for scanning in FIG. 8 can be omitted. This is the same as when the fan beam FB is irradiated (see FIG. 6).

<Signal input, reference level storage, and level judgment>
First, signal input, reference level storage, and level determination will be described. In FIG. 8, after the start of the vehicle detection method, the reference level BL is set or updated in step S01. The reference level BL includes a road surface reflection signal level RL and a device noise level NL. In step S01, the setting or updating of the reference level BL includes acquisition and storage of the road surface reflection signal level RL in step S02 and acquisition and storage of the device noise level NL in step S03.

  The road surface reflection signal level RL is a reflection signal detected by transmitting and receiving radio waves by the sensor device 2 of the vehicle detection device 1, and is a reflection signal from the road surface. The device noise level NL is a signal detected from the sensor device 2 of the vehicle detection device 1 and is caused by noise components such as dark current in the sensor device 2. Therefore, the device noise level NL is inherent in the road surface reflection signal level RL.

  The road surface reflection signal level RL is detected by the sensor control unit 23 switching the transmission / reception switch 213 at a timing when a reflected radio wave is obtained only from the road 15 and the intersection 13 without the vehicle 7 in the detection range DA of the sensor unit 21. Can be obtained. The obtained road surface reflection signal level RL is stored in the reference level storage unit 313 (step S02). Referring to FIG. 1, the detection of road surface reflection signal level RL may be performed over the entire detection range DA from position P1 to position P4, or may be limited to a necessary range.

  FIG. 9 is a graph showing the road surface reflection signal level RL and the equipment noise level NL. The vertical axis indicates the signal intensity of the received radio wave, and the horizontal axis indicates positions P0 to P4 (see FIG. 1). As shown in FIG. 9, different road surface reflection signal levels RL can be obtained at positions P1 to P4. The road surface reflection signal level RL decreases as the position P1 moves to the position P4. This is because, as the position in the detection range DA is farther from the sensor device 2, the radio wave is diffused by being attenuated with distance, and the signal intensity of the received radio wave is weakened. It should be noted that different road surface reflection signal levels RL can be obtained at each position orthogonal to the line connecting the position P1 and the position P4, that is, at each position in the width direction of the road.

  The device noise level NL is measured with the sensor control unit 23 disconnecting the antenna connection switch 214. In a state where the sensor device 2 is separated from the antenna 211 by a circuit, the sensor device 2 acquires a signal as the device noise level NL. The obtained device noise level NL is stored in the reference level storage unit 313 (step S03). Therefore, as shown in FIG. 9, substantially the same device noise level NL can be obtained regardless of the positions P1 to P4 on the road. The device noise level NL is equal to the output of the vehicle detection device 1 when the transmitted radio wave cannot be received without being reflected (when there is no reflection). Therefore, the device noise level NL is also referred to as a non-reflection level, and may occur even when a reflected wave cannot be received by the sensor unit 21 due to specular reflection of an object.

  In addition, even when the sensor unit 21 is scanned as shown in FIG. 6 and the fan beam FB is irradiated, as shown in FIG. The signal level RL and the device noise level NL are obtained similarly.

  Note that step S01 is performed at any time prior to the subsequent steps. This is because the strength of radio waves detected varies depending on the weather and various environments such as weather and temperature, and therefore, by using the latest reference level BL, the detection accuracy of reflected signals described below is improved. However, regardless of this, the past road surface reflection signal level RL and the equipment noise level NL may be used without performing step S01 for the purpose of improving the processing speed.

  Next, in step S04, transmission and reception of radio waves are performed for the entire area in the road extending direction where the sensor unit 21 can transmit radio waves. In the case of scanning in the example of FIG. 1 (see FIG. 6), transmission and reception of radio waves are performed within a range of one fan beam extending from position P1 to position P4. In addition, when a fixed sensor unit 21 that does not scan is installed for each lane of travel (see FIG. 7), all fan beams may be irradiated simultaneously.

  The signal input control unit 311 outputs a control signal to the sensor control unit 23. In response to this control signal, the sensor control unit 23 switches the antenna connection switch 214 to the antenna connection state, and appropriately switches the transmission / reception switch 213 to transmit / receive radio waves. As a result, radio waves reflected by the oncoming vehicle 77, the road 15, the intersection 13, and the like in one fan beam in the detection range DA are received via the antenna 211. The received radio wave becomes a detection signal DS via the antenna connection switch 214, the reception side amplifier 215, and the reception side A / D converter 216, and then passes through the sensor control unit 23 to control the signal input of the signal detection unit 31 of the signal processing device 3. The signal is input to the level determination unit 312 via the unit 311 and the reference level storage unit 313. The signal input control unit 311 outputs an apparatus noise level NL acquisition or update signal to the reference level storage unit 313.

  In step S05, the vehicle reflection signal VS related to the vehicle 7 to be noted is specified from the detection signal DS. A method for specifying a signal from the vehicle 7 to be noted will be described with reference to FIGS. 10 and 11. FIG. 10 shows an example of the detection signal DS obtained in step S04 on the coordinates formed by the vertical axis between the positions P1 and P4 and the horizontal axis of time. FIG. 11 shows another example of the detection signal DS obtained in step S04 on the coordinates formed by the vertical axis between the positions P1 and P4 and the horizontal axis of time.

  Referring to FIG. 10, VS1 represents a vehicle reflection signal from a certain vehicle, and VS2 represents a vehicle reflection signal from a vehicle different from VS1. A broken line between the vehicle reflection signal VS2 and the vehicle reflection signal VS1 indicates a reflection signal from the road surface detected from the rear portion of the vehicle corresponding to VS2. A broken line on the right side of the vehicle reflection signal VS1 indicates a reflection signal from the road surface detected from the rear portion of the vehicle corresponding to VS1. The intensity of the reflected signal indicated by the broken line corresponds to the road surface reflected signal level RL.

  Referring to FIG. 10, vehicle reflection signal VS2 changes from a distant position to a close position with the passage of time. Therefore, it can be seen that the vehicle corresponding to the vehicle reflection signal VS2 is the oncoming vehicle 77 traveling toward the installation position of the sensor unit 21. Here, since the distance between the position P4 and the position P1 is, for example, 120 m, the speed of the vehicle corresponding to the vehicle reflection signal VS2 can be determined from the distance that the vehicle reflection signal VS2 has changed with respect to the elapsed time. The vehicle corresponding to the vehicle reflection signal VS1 is also traveling toward the installation position of the sensor unit 21, and appears on the coordinate plane at a time later than the vehicle reflection signal VS2. It turns out that it is 78. Here, the inter-vehicle distance between the oncoming vehicle 77 and the rear oncoming vehicle 78 can also be specified as the distance between the positions where both exist at a predetermined time. As described above, the positional relationship and the vehicle speed between the oncoming vehicle 77 and the rear oncoming vehicle 78 can be understood from the time change of the vehicle reflection signals VS1 and VS2. If there is a signal that does not change over time, it is a fixed object such as a guardrail or a utility pole. Moreover, although it changes with progress of time, it can be judged that the thing which is very weak or slow is a pedestrian or a bicycle. Here, since the oncoming vehicle 77 and the rear oncoming vehicle 78 are handled as the vehicle 7 to be noted, the detection signal DS including the vehicle reflection signals VS1 and VS2 is a signal handled in the subsequent processing.

  Referring to FIG. 11, VS5 to VS9 show vehicle reflection signals from different vehicles 7, respectively. From the time change of the vehicle reflection signals VS5, VS6, and VS7, it can be seen that the vehicle reflection signals VS5, VS6, and VS7 are oncoming vehicles 77. Here, the oncoming vehicle corresponding to the vehicle reflection signal VS5 exists in a horizontal line state from about 12:45 to 20 seconds about 80 m ahead of the sensor unit 21, and after about 10 seconds, there is no transition about 40 meters ahead. Therefore, the car stopped about 20m from 12:45 in front of about 80m before the oncoming car corresponding to the vehicle reflection signal VS6 started to move. It can be inferred that the detection has gone out of the detection range DA.

  The vehicle reflection signals VS8 and VS9 are moving away from the sensor unit 21 with time. Accordingly, the vehicle reflection signals VS8 and VS9 do not represent oncoming vehicles, and no attention is paid here. The vehicle reflection signals VS6 and VS7 indicate the oncoming vehicle 77 and the rear oncoming vehicle 78, as described with reference to FIG.

  As described above, in the example of FIG. 11, the detection signal DS including the vehicle reflection signals VS6 and VS7 is set as a target of processing to be described later as a signal of the vehicle 7 to be noted.

  In addition, the range shown by hatching in the upper left of the coordinates in FIG. 11 is a range (outside the visibility distance) that is difficult to observe visually by using a visibility meter from the position of the sensor unit 21 in a certain road condition. When the visibility is poor due to changes in the rain, fog, and sunlight conditions, the presence of the oncoming vehicle 77 may finally be determined with the naked eye. However, according to the embodiment, it is possible to receive the vehicle reflection signals VS6, VS7 and VS8 from the range shown by hatching in FIG. Even in such a poor field of view, the driver of the right turn car 73 can notice the presence, number, arrangement, etc. of oncoming vehicles faster than the naked eye by using the vehicle detection device of the embodiment.

  Next, in step S06, the level determination unit 312 determines the road surface reflection signal level RL corresponding to the vehicle reflection signal corresponding to the vehicle to be noted, that is, the road surface reflection at the position in the width direction of the road where the vehicle reflection signal exists. The road surface reflection signal level RL, which is the signal level RL and obtained in step S02 immediately before acquisition of the vehicle reflection signal, is read from the reference level storage unit 313. Further, the device noise level NL obtained in step S02 immediately before acquisition of the vehicle reflection signal is also read from the reference level storage unit 313. Note that the road surface reflection signal level RL and the equipment noise level NL may be read at intervals scanned along the road width direction in the range from the position P1 to the position P4 in the detection range DA.

  FIG. 12 is an example of a graph of the intensity of the detection signal DS from the oncoming vehicle 77 and the backward oncoming vehicle 78 that are traveling on the road 15 and the intersection 13 in the entire detection range DA. In FIG. 12, the road surface reflection signal level RL and the device noise level NL shown in FIG. 9 are also displayed. Conceptually, the information shown in FIG.

  In FIG. 12, the vertical axis represents signal intensity, and the horizontal axis represents position P0 to position P4 (see FIG. 1). The vehicle reflection signal VS1 and the vehicle reflection signal VS2 correspond to the vehicle reflection signal VS1 and the vehicle reflection signal VS2 in FIG. 10, respectively. The vehicle reflection signal VS <b> 2 indicates a vehicle reflection signal from the oncoming vehicle 77 having a vehicle height higher than that of the oncoming vehicle 78 of the vehicle reflection signal VS <b> 1. This is because the larger the vehicle, the larger the reflected area of the radio wave, the stronger the reflected signal intensity, and the higher the vehicle height, the lower the radio wave that is transmitted, reflected, and received. This is because the signal strength becomes stronger because it is less.

  FIG. 13 is an example of the detection signal DS obtained in a predetermined road width range in the detection range DA. In FIG. 13, for comparison with the detection signal DS, the detection signal VS1, the detection signal VS2, the road surface reflection signal level RL, and the device noise level NL of FIG. 12 are also shown. That is, the detection signal VS1 from the oncoming vehicle 77, the detection signal VS2 from the rear oncoming vehicle 78, and the road surface reflection signal level RL are indicated by broken lines, and the device noise level NL is indicated by a one-dot chain line. Hereinafter, an arbitrary position between the positions P1 to P4 will be described using a variable PX.

  In step S07, the level determination unit 312 defines the horizontal position closest to the sensor unit 21 as PX = P1. In step S08, the level determination unit 312 determines whether PX exceeds P4 (see FIG. 1). If PX exceeds P4 (YES), the process proceeds to step S20. If PX does not exceed P4 (NO), the process proceeds to step S09. In step S09, the level determination unit 312 determines a shadow area. The level determination unit 312 determines whether the detection signal intensity DSI is equal to or lower than the device noise level NL at the position PX. In addition, since the intensity of the equipment noise level NL is inherent in the detection signal intensity DSI, if the detection signal intensity DSI is equal to or less than the equipment noise level NL, it means that the vehicle is not detected from the sensor unit 21. Here, when acquiring in step S03, the maximum value of the fluctuation included in the received signal is set as the device noise level NL.

  Referring to FIG. 13, for example, when position PX is between position P21 and position P20, detection signal intensity DSI is equal to or less than device noise level NL (YES), and therefore, between position P21 and position P20 is determined to be a shadow region. (Step S14), the process proceeds to Step S20. For example, when the position PX is between the position P20 and the position P32, the detection signal intensity DSI is not less than or equal to the device noise level NL. When the loop processing from step S08 to step S19 is completed, it is determined that the area is a shadow area from position P0 to position P1, from position P21 to position P21, and from position P31 to position P30. (Step S14).

  In step S10, the level determination unit 312 determines the road surface reflection area. The level determination unit 312 determines whether the detection signal intensity DSI is equal to the road surface reflection level RL at the position PX. Referring to FIG. 13, for example, when the position PX is between the position P21 and the position P22, the detection signal intensity DSI exceeds the road surface reflection level RL. Therefore, the position P21 to the position P20 are not determined as the road surface reflection area (NO ), The process proceeds to step S11. For example, when the position PX is between the position P20 and the position P32, the detection signal intensity DSI is equal to or less than the road surface reflection level RL (YES), so the position P20 to the position P32 is determined as a road surface reflection region (step S15), and step The process proceeds to S20. In consideration of the error, when the detection signal intensity DSI exceeds the device noise level NL but is less than or equal to the road surface reflection level RL, the road surface reflection region may be determined, and otherwise, the road surface reflection region may not be determined ( NL <DSI ≦ RL). When the loop processing from step S08 to step S19 is completed, it is determined that the area between the position P1 and position P22, the position between position P20 and position P32, and the position between position P30 and position P4 is a road surface reflection area. (Step S15).

  In step S11, the level determination unit 312 determines the reflection from the small vehicle. The level determination unit 312 determines whether the detection signal intensity DSI is, for example, the vehicle reflection signal intensity VSI1 or less at the position PX. Referring to FIG. 13, for example, when position PX is between position P32 and position P31, detection signal intensity DSI is equal to or lower than vehicle reflection signal intensity VSI1 (YES), so that there is a small size between position P32 and position P31. It is determined that there is reflection from the vehicle (step S16), and the process proceeds to step S20. For example, since the detection signal intensity DSI is not less than or equal to the vehicle reflection signal intensity VSI1 when the position PX is between the position P22 and the position P21, it is determined that there is no reflection from the small vehicle from the position P22 to the position P21 (NO). The process proceeds to step S12. In consideration of the error, when the detection signal intensity DSI exceeds the road surface reflection level RL but is less than or equal to the intensity VSI1 of the vehicle reflection signal VS1 (RL <DSI ≦ VSI1), it is determined that the reflection is from a small vehicle. Sometimes, it may not be determined that the light is reflected from a small vehicle. When the loop processing from step S08 to step S19 is completed, it is determined that there is reflection from the small vehicle between position P31 and position P32 (step S16).

  In step S12, the level determination unit 312 determines reflection from a large vehicle. The level determination unit 312 determines whether the detection signal intensity DSI is, for example, the vehicle reflection signal intensity VSI2 or less at the position PX. Referring to FIG. 13, for example, when position PX is between position P22 and position P21, detection signal intensity DSI is equal to or lower than vehicle reflection signal intensity VSI2 (YES), so that there is a large area between position P22 and position P21. It is determined that there is reflection from the vehicle (step S17), and the process proceeds to step S20. For example, when the position PX is between the position P32 and the position P31, the detection signal intensity DSI is less than or equal to the vehicle reflection signal intensity VSI2, but is also less than or equal to the vehicle reflection signal intensity VSI1, so between the position P32 and the position P31 It is determined that there is no reflection from the large vehicle (NO), and the process proceeds to step S13. In consideration of the error, when the detection signal intensity DSI exceeds the vehicle reflection signal intensity VSI1 but is not more than the vehicle reflection signal intensity VSI2 (VSI1 <DSI ≦ VSI2), it is determined that the reflection is from a large vehicle. You may make it not determine with the reflection from a large vehicle. When the loop processing from step S08 to step S19 is completed, it is determined that there is reflection from the large vehicle between position P21 and position P22 (step S17).

  Although only two types of vehicle reflection signal intensity VSI1 of vehicle reflection signal VS1 and vehicle reflection signal intensity VSI2 of vehicle reflection signal VS2 have been described, the vehicle reflection signal intensity may be three or more types, Can be set as appropriate. The set plurality of vehicle reflection signal intensities may be used in the level determination unit 312 by appropriately storing them in a table or the like.

  In step S13, the level determination unit 312 determines an exception. The level determination unit 312 determines whether the detection signal intensity DSI is greater than the vehicle reflection signal intensity VSI2 at the position PX. For example, when the detection signal intensity DSI is greater than the vehicle reflection signal intensity VSI2, the right turn driver is a reflective object that can be noticed visually, so the process proceeds to steps S19 and S20 as exception determination (step S18).

  Note that the shadow region, road surface reflection region, small vehicle, large vehicle, and exception determination from step S14 to step S18 are primary determinations. The determination accuracy may be improved by using the determination of the large vehicle and the small vehicle (steps S35 and S36) described later with reference to FIG. 14 and the primary determination (steps S14 to S18).

  In step S19, the arbitrary position PX is moved by a predetermined interval PG to be a new position PX. As long as the new position PX does not exceed the position P4 farthest from the sensor unit 21 within the predetermined range in the determination in step S08, the shadow area, the road surface reflection area, the small vehicle, the large vehicle, the exception for the new position PX. Is determined (step S09 to step S18).

  In step S20, for each position PX, the level determination unit 312 causes the determination information storage unit 314 to store the result data determined to be any type of shadow region, road surface reflection region, small vehicle, large vehicle, or exception. . At this time, the level determination unit 312 also determines each position (position where the radio wave is reflected), the reflected signal intensity at each position, the horizontal distance from the position P0 at that position, the vertical distance from the position to the road surface, and the like. The data is stored in the storage unit 314. When the sensor unit 21 scans and transmits and receives radio waves, the level determination unit 312 also acquires the movable angle (that is, the azimuth angle) of the sensor unit 21 in the road width direction from the sensor control unit 23 and the like. The determination information is stored in the determination information storage unit 314 in association with the determined result data. Thereby, as shown in FIG. 13, for example, detailed data relating to the transition of the intensity of the detection signal DS is stored for a part of the detection range DA. The transition of the reflected signal intensity can be simplified to at least four stages. In addition, in order to obtain | require a horizontal distance from the distance from the sensor part 21 to the vehicle 7 or the road 15, it can calculate using the trigonometric function of the angle which transmitted / received the electromagnetic wave in the sensor part 21. FIG. For example, the sine (sine) of the angle at which radio waves are transmitted and received by the sensor unit 21 with respect to the vertical line is integrated with the distance from the sensor unit 21 to the vehicle 7 or the road 15 acquired by the sensor unit 21 by a known method. . In this way, the horizontal distance from the sensor unit 2 to the detection target is obtained. Further, in order to obtain the vertical distance from the distance from the sensor unit 21 to the vehicle 7 or the road 15, it can be calculated using a trigonometric function of the angle at which the sensor unit 21 transmits and receives radio waves. For example, the cosine of the angle at which radio waves are transmitted and received by the sensor unit 21 with respect to the vertical line is integrated with the distance from the sensor unit 21 to the vehicle 7 or the road 15 acquired by the sensor unit 21 by a known method. . In this way, the vertical distance from the sensor unit 21 to the detection target is obtained.

  In step S21, a rising point and a falling point of the detection signal strength DSI are extracted. In step S22, it is determined whether the type of detection signal in the previous scan has changed. Here, the type of detection signal refers to any type of shadow area, road surface reflection area, small vehicle, large vehicle, or exception. If the type of the detection signal has not changed in step S22 (YES), the process proceeds to step S24. If the type of the detection signal has changed in step S22 (YES), the detected position PX and the detection signal intensity DSI at that position PX are determined for each type of the detection signal in step S23. The determination information storage unit 314 stores each position PX detected by the level determination unit 312 and the detection signal intensity DSI at the position PX for each type of the detection signal (step S23). Thereafter, the process proceeds to step S24.

  In step S24, it is determined whether radio waves have been transmitted and received in the entire range on the detection range DA by scanning. If the scanning is complete and the data such as the intensity of the detection signal DS is acquired by the level determination unit 312 and stored in the determination information storage unit 314 (YES), the process proceeds to step S26. When all the ranges on the detection range DA have not been scanned (NO), the process proceeds to step S25. In step S25, for example, the sensor control unit 23 changes the radio wave transmission / reception direction to the width direction of the road by scanning the sensor unit 21 to a range not yet scanned on the detection range DA.

  In step S26, for example, it is determined whether the signal input control unit 311 or the sensor control unit 23 has received a detection end instruction. If an end instruction has been given (YES), the process proceeds to step S28. In step S28, each detected position PX and the detected signal intensity DSI at the position PX are output from the determination information storage unit 314 to the tracking unit 32. Thereafter, the process ends.

  If no termination instruction is given in step S26 (NO), the process proceeds to step S27.

  In step S27, for example, the sensor control unit 23 determines whether a predetermined time has elapsed since the previous detection of the entire detection range DA. If a predetermined time has elapsed since the previous detection of the entire detection range DA (YES), the process returns to step S01 and the detection process is performed again. If the predetermined time has not elapsed since the previous detection of the entire detection range DA (NO), the process returns to step S26 and waits for the elapse of the predetermined time.

  In step S26, it is determined whether the signal input control unit 311 or the sensor control unit 23 has received a detection end instruction. If the entire detection range DA is scanned without waiting for this determination (step S24). The process may be terminated at YES).

  As described above, the determination information storage unit 314 has each position (a position where the radio wave is reflected) and a shadow at each position for every range on the detection range DA for each position PX moved by the predetermined interval PG and for each scan. Area, road surface reflection area, small vehicle, large vehicle, result data determined to be any kind of exception, reflected signal intensity at each position, horizontal distance from position P0 at each position, vertical from that position to the road surface The distance, the movable angle (azimuth angle) of the sensor unit 21 in the width direction of the road, and the like can also be stored and accumulated.

<Calculation of vehicle height and length, determination of vehicle shape, and determination of vehicle type>
Next, calculation of the vehicle length and height, determination of the shape of the vehicle, and determination of the vehicle type will be described. FIG. 14 is a flowchart of calculation of the vehicle length and vehicle height, determination of the shape of the vehicle, and determination of the vehicle type.

  First, in step S31, the tracking unit 32 of the signal detection unit 31 of the signal processing device 3 specifies, for example, the oncoming vehicle 77 closest to the sensor unit 21 from the data stored in the determination information storage unit 314. The data regarding the specified oncoming vehicle 77 is specified by labeling, for example. Information on the identified oncoming vehicle 77 is determined at predetermined time intervals by the level determination unit 312 in steps S20 to S23 of FIG. 8 and is also included in a plurality of data sets stored in the determination information storage unit 314. Therefore, the tracking unit 32 specifies and holds all the data related to the oncoming vehicle 77 in the data for every predetermined time by using a known vehicle tracking method. Alternatively, the data stored in the determination information storage unit 314 is overwritten by the tracking unit 32 with a part of the data specified by labeling or the like, and stored in the determination information storage unit 314.

  The tracking unit 32 specifies the oncoming vehicle 77 to be tracked instead of specifying the oncoming vehicle 77 closest to the sensor unit 21 from the data stored in the determination information storage unit 314, and stores data related thereto. You may instruct | indicate with respect to the signal input control part 311 to acquire. In this case, the signal input control unit 311 instructs the sensor control unit 23 of the sensor device 2 to acquire data related to the oncoming vehicle 77 to be tracked. Further, the sensor control unit 23 instructs the sensor unit 21 of the sensor device 2 to acquire data related to the oncoming vehicle 77 to be tracked. By doing in this way, the data of the oncoming vehicle 77 to be tracked can be preferentially acquired over time and the vehicle shape can be quickly determined.

  In step S <b> 32, the vehicle length height calculation unit 315 reads data at a certain predetermined time and data at the next adjacent predetermined time among the data relating to the identified oncoming vehicle 77, for example, in the oldest order. The tracking unit 32 determines the calculation priority according to the current traveling position of the oncoming vehicle 77 on the road, and determines the old order, new order, old order from the middle, new order from the middle, You may read in order.

  Note that the processing from steps S20 to S23 in FIG. 8 may be included in the processing in step S32 in FIG.

  In step S33, the vehicle length vehicle height calculation unit 315 calculates the detection length c and the detection height h of the vehicle body of the oncoming vehicle 77 as described below. FIG. 15 is an explanatory diagram for explaining a method of calculating the detection length and the detection height of the vehicle body.

  In FIG. 15, H indicates the height of the sensor unit 21 from the road. θ is a downward angle of the sensor unit 21 with respect to the column 16. h is the detection height. In FIG. 15, a cargo vehicle with a cargo bed is illustrated as an oncoming vehicle 77. In the case of such a truck, the front half of the vehicle body corresponds to the passenger cabin portion of the vehicle body, and the passenger cabin portion is higher than the loading platform. For example, the detection length c is the position P22 on the road 15 corresponding to the rising point from the road surface reflection signal level RL closest to the sensor unit 21 to the vehicle reflection signal intensity VSI2 and the vehicle reflection signal intensity in the detection signal DS in FIG. It can be acquired as the distance from the position P21 on the road 15 corresponding to the falling point from the VSI 2 to the device noise level NL. F indicates the horizontal distance from the position where the vertical line of the sensor unit 21 intersects the road 15 to the front end of the vehicle body. L is detected from the position where the vertical line of the sensor unit 21 intersects the road 15 by radio waves transmitted and received from the sensor unit 21 in the vicinity of the rear end of the vehicle body or the upper half of the vehicle body (the cabin part). The horizontal distance to the detection position of the road 15 is shown. B is between the rear end portion of the vehicle body or the front half of the vehicle body and the detection position of the road 15 detected by the radio waves transmitted and received in proximity to the rear end portion of the upper part of the front half of the vehicle body from the sensor unit 21. Indicates the horizontal distance.

  Referring to FIG. 15, the following relationship is established.

  Therefore, Equation (3) is derived from Equation (1) and Equation (2).

  Here, since the loading platform is behind the passenger compartment which is the front half of the vehicle body, the positional relationship in FIG. 15 is irrelevant to the detection, and a box-type vehicle having a vehicle height h and a vehicle length c (a vehicle without a loading platform) is detected. The same detection result is shown.

The vehicle length vehicle height calculation unit 315 performs calculation using data at two different predetermined times in the specified data. Note that the horizontal distance F from the sensor unit 21 to the front end of the vehicle body at an old predetermined time, the horizontal distance L from the sensor unit 21 to the detection position on the road by the sensor unit 21, and the road by the sensor unit 21 from the rear end part of the front half of the vehicle body The horizontal distance B to the upper detection position is assumed to be F ₀ , L ₀ and B ₀ , respectively. The horizontal distance F from the sensor unit 21 to the front end of the vehicle body at a new predetermined time, the horizontal distance L from the sensor unit 21 to the detection position on the road by the sensor unit 21, and the detection of the sensor unit on the road from the rear end portion of the front half of the vehicle body The horizontal distance B to the position is F ₁ , L ₁ , and B ₁ , respectively. The height H, the detection height h, and the detection length c from the road of the sensor unit 21 are equal between the old predetermined time and the new predetermined time.

  However, Formula (2) will be described since there are points to be considered when determining the horizontal distance L from the sensor unit 21 to the detection position on the road by the sensor unit 21. FIG. 16 shows an example of a detection signal DS obtained in a predetermined range in the detection range DA within a predetermined time, and shows another example different from FIG. Referring to FIG. 13, a horizontal distance L from the sensor unit 21 to the detection position on the road of the sensor unit 21 is between the position P0 and the position P20. Since it is determined that the area between the position P20 and the position P32 is a road surface reflection area (step S15), when looking down from the sensor unit 21, the shadow of the oncoming vehicle 77 is behind the position P20 (in the direction toward the position P4). It can be seen that the road surface behind appears. On the other hand, referring to FIG. 16, since the device noise level NL is detected in a part of the detection signal from position P32 to position P21, the detection signal of the road surface reflection area is detected by the detection signal DS of the backward oncoming vehicle 78. The position where the rear road surface appears from behind the oncoming vehicle 77 cannot be specified. Therefore, in the case of FIG. 16, it is unclear whether the position between the position P0 and the position corresponds to the horizontal distance L from the sensor unit 21 to the detection position on the road by the sensor unit 21. Therefore, the signal strength waveform data as shown in FIG. 16 may not be preferentially used in the calculation in the vehicle length vehicle height calculation unit 315 described later.

  Next, Equation (4) is derived from Equation (3).

  In equation (4), when c is moved to the leftmost side, equation (5) is obtained.

Here, L ₀ , L ₁ , F ₀ , and F ₁ are included in the data read from the determination information storage unit 314 to the vehicle length vehicle height calculation unit 315 as the horizontal distance, so the detection length c Is uniquely determined. The detected height h is uniquely determined by substituting the length c of the front half of the vehicle body thus obtained together with L ₀ , F ₀ , and H into Equation (4).

  In this way, the length of the front half of the vehicle body is obtained as the detection length c, and the height of the front half of the vehicle body is obtained as the detection height h. As described above, for example, for a box-type vehicle (a vehicle without a loading platform), the vehicle length is similarly obtained as the detection length c and the vehicle height is obtained as the detection height h.

  Next, in step S34, the vehicle shape vehicle type determination unit 317 determines the detected height h, which is the height of the vehicle body front half or the vehicle height, obtained in this way, by comparing with a separately prepared height discrimination value. For example, as an example of the discriminant value, it may be determined that the vehicle is a large vehicle when the detection length exceeds 4.7 m or the detection height exceeds 2.0 m according to the regulations of the vehicle law. If the detected height h is greater than the height determination value (YES), it is determined that the oncoming vehicle 77 is a large vehicle (step S35). On the other hand, if the detected height h is less than or equal to the height determination value (NO), it is determined that the oncoming vehicle 77 is a small vehicle (step S36).

  At this time, if the vehicle reflected signal intensity VSI from the oncoming vehicle 77 has reached the intensity of the vehicle reflected signal VS2 as shown in FIG. 13, it can be supplementarily determined that the vehicle is a large vehicle. Similarly, when the vehicle reflected signal intensity VSI from the oncoming vehicle 77 reaches the intensity of the vehicle reflected signal VS1 as shown in FIG. 13 but does not reach the intensity of the vehicle reflected signal VS2, the small vehicle is auxiliary. Can be judged.

  In addition, in combination with the vehicle type obtained from the vehicle number recognition unit 33, the vehicle type in steps S35 and S36 may be determined. Thereby, tracking can be made highly accurate.

  Next, in step S37, it is determined whether the vehicle is a truck type or a bus type among large vehicles. The truck type refers to a vehicle having a shape such as a cargo vehicle with a loading platform shown in FIG. The bus type means, for example, a box-type vehicle whose roof height does not change.

  FIG. 17 shows the detected height h obtained by calculating step S32 from position P1 to position P4, that is, every adjacent predetermined time from the position of the detection range DA to the depth of about 120 m with 0 m as the position on the sensor unit 21 side. And a change in the detection length c corresponding to the position of the oncoming vehicle 77 on the road. Here, it is assumed that the oncoming vehicle 77 has a track type shape as shown in the lower right of FIG. That is, the height of the front half of the vehicle body is 4 m, the length of the front half of the vehicle body is 1 m, the height of the loading platform is 2 m, and the total length is 12 m.

  Referring to the graph of FIG. 17, according to the calculation result in the vehicle length vehicle height calculation unit 315 of the vehicle shape feature determination unit 319 in step S <b> 33, the detected height is about 120 m to about 30 m from the position P <b> 0 as a starting point. h and the detection length c are substantially constant. Here, the detection height h is about 4 m, and the detection length c is about 1 m. As described with reference to FIG. 15, at the position away from the sensor unit 21, the loading platform is behind the front half of the vehicle body, and the transmitted radio waves are not reflected by the loading platform. This is because the light is reflected by the sensor unit 21 and received by the sensor unit 21.

  However, the detection height h decreases from 4 m to 2 m and the detection length c increases from 1 m to 12 m from about 30 m to about 15 m from the position P1. This is due to the fact that the oncoming vehicle 77 is sufficiently close to the sensor unit 21 and thus the loading platform is gradually opened from the shadow of the cabin. That is, the radio wave transmitted from the sensor unit 21 starts to be reflected by the cargo bed unit and is received by the sensor unit 21.

  Further, the detection height h is substantially constant at 2 m and the detection length c is substantially constant at 12 m from the position P1 to a position from about 15 m to about 0 m. This is because the oncoming vehicle 77 has sufficiently approached the sensor unit 21, so that the cargo bed unit is completely opened from behind the passenger compartment. That is, the radio wave transmitted from the sensor unit 21 is reflected by the entire vehicle body of the oncoming vehicle 77 and received by the sensor unit 21. Here, the detection height h and the detection length c indicate the vehicle height and the vehicle length from about 15 m ahead of the position P1 to about 0 m. Depending on the position on the road and the vehicle type, the vehicle length vehicle height calculation unit 315 calculates the height of the passenger compartment, the vehicle height, or a mixed state thereof, and the length of the passenger compartment, the vehicle length, or a mixed state thereof. The calculation is performed as the detection height h and the detection length c.

  On the other hand, when the oncoming vehicle 77 is a bus type, the detected height h and the detected length c do not change in the graph shown in FIG. Therefore, in step S37, if there is a change (YES) depending on whether the detection height h has changed at a predetermined position or the detection length c has changed at a predetermined position, the track type is determined. (Step S38) If there is no change (NO), the bus type is determined (Step S39). In addition, for the sake of simplicity, the description of the determination of the type of the large vehicle has been made with reference to the truck type and the bus type. However, the change in the detection height h at a predetermined position or the change in the detection length c at a predetermined position. There are various variations, and it is possible to determine more types of vehicles by categorizing the state of change.

  Next, in step S40, it is determined whether it is a passenger car type among small cars or a box type. The passenger car type refers to a vehicle shaped like a sedan with a trunk, for example. The box type means, for example, a box type vehicle whose roof height does not change.

  As described with reference to FIG. 17, in the passenger car type in which the height of the upper portion changes similarly to the truck type, the detection height h changes at a predetermined position, or the detection length c changes at the predetermined position. On the other hand, when the oncoming vehicle 77 is a box type, there is no increase / decrease change in the detection height h and the detection length c in the graph shown in FIG.

  Therefore, in step S40, if there is a change (YES) depending on whether the detection height h has changed at a predetermined position or the detection length c has changed at a predetermined position, it is determined as a passenger car type. (Step S41) When there is no change (NO), it is determined as a box type (Step S42). In addition, although the passenger car type and the box type have been described for the sake of simplicity as the description of the vehicle type determination of the small car, the change in the detection height h at a predetermined position or the change in the detection length c at the predetermined position is described. There are various variations, and more vehicle types can be determined by categorizing the state of change.

  In step S43, it is determined whether or not the data of all adjacent two predetermined times regarding the oncoming vehicle 77 to be determined is read in step S32 and set as the calculation target in step S33.

  While performing the vehicle shape calculation and vehicle type determination processing from step S31 to step S42 shown in FIG. 14, the detection signal DS determination processing from step S01 to step S27 shown in FIG. 8 may be performed. In this case, the data obtained by the determination process of the detection signal DS may be used in the calculation of the vehicle shape and the determination process of the vehicle type as needed.

  In addition, with reference to FIG. 17, the position of the vehicle for every predetermined time is graphed. By using this data, the vehicle shape vehicle type determination unit 317 can detect the speed of the oncoming vehicle 77 and the transition of the speed over time. As described above, the positional relationship between the oncoming vehicle 77 and the rear oncoming vehicle 78 and the vehicle speed can be detected from the time change of the vehicle reflection signals VS1 and VS2. Alternatively, the vehicle speed can also be detected from the sensor unit 21 that uses a radar speed detection function such as the Doppler effect.

  The determination information such as the vehicle length, vehicle height, vehicle speed, vehicle shape, and / or vehicle type obtained in this manner is sent to the communication center 5 via the communication unit 318, and the communication center 5 is on the road. Such information is appropriately provided to the information providing apparatus 6, and the right turn car 73 can acquire the information from the information providing apparatus 6 and use it for safe driving.

  Thus, according to the vehicle detection device 1, it is possible to determine the characteristics of the shape of the traveling vehicle. Still further, according to the vehicle detection device 1, it is possible to realize the tracking of the positions of a plurality of vehicles and the determination of the characteristics of the shapes of these vehicles with a single device. Considering that the technology described in the prior art documents mentioned above is going to do the same thing, multiple radars in the vicinity of the intersection of a radar for position tracking and a radar for the purpose of distinguishing between large and small vehicles It will be necessary to provide in. However, according to the vehicle detection apparatus 1, since the position tracking and the shape feature determination can be performed by one apparatus, the apparatus cost can be reduced. In addition, the installation time can be reduced. In addition, since the number of devices can be reduced, the number of man-hours and costs for maintenance during operation can be reduced.

<When using an optical camera as the sensor unit 21>
In the vehicle detection device and the vehicle detection method according to the embodiment, the vehicle 7 is detected by transmitting and receiving radio waves. However, the optical camera is used as the sensor unit 21 and the same function as in the embodiment is used. Can be realized. In this case, since optical image information is handled, there is no transmission such as radio waves, and calculation of the vehicle position, vehicle length, vehicle height, and vehicle speed, vehicle shape determination, and vehicle type determination are performed only by image reception. It can be carried out.

  Various known cameras can be used as the optical camera. For example, the range of the shooting depth distance from 30 m to 120 m can be set within the depth of field. Since the detection signal DS is light reflected from an object, it is desirable that the number of pixels of the image sensor is large and the light receiving surface is wide in order to obtain sufficient signal intensity. Assuming that the amount of natural light is not large, such as at dusk, it is desirable to use a lens with a large F value obtained by dividing the focal length of the lens by the effective aperture. In addition, you may use together the illumination which can illuminate on the opposite lane from the position P1 to P4 for the detection at nighttime.

  In addition, referring to FIG. 1, when the distance from the position P0 to the position P1 is 30 m, if the opposite lane width is 10 m and a lens having an angle of view of about 40 degrees or more is used, the intersection entrance The opposite lane width of 10 m can be completely within the field of view. When using a lens with a narrower angle of view, use a known device that can scan the optical camera in the same way as in the embodiment, or parallel multiple optical cameras in the width direction of the road. Thus, it is possible to secure a field of view for acquiring the detection signal.

  For the convenience of image processing, it is desirable to have a function capable of digitizing a recorded image. Since light is one type of radio wave, the processing described in the embodiment can be similarly realized even when an optical camera is used.

  When an optical camera is used, the detection length c and the detection height h can be obtained separately using a known image analysis method. In this case, for example, if a surveying bar having the same scale is placed on the left and right and front and rear in the imaging range, an accurate detection height h can be obtained on the perspective view. The detection height h obtained in this way is used to check the accuracy of the calculation result of the detection height h obtained by substituting into the above-described equation (4). It can be used to confirm accuracy.

  In particular, in the case of an optical camera, the upper left screen in FIG. 3 can be provided to the right turn car 73 or the like via the communication unit 318, the communication center 5, and the information providing device 6 as they are.

  Although the embodiment has been described in detail above, the present invention is not limited to this embodiment, and various modifications and changes other than the above embodiment can be made within the scope described in the claims.

Further, the following supplementary notes are disclosed with respect to the above embodiments.
(Appendix 1)
A sensor unit that is installed above the road and detects the road and the vehicle to be determined as a detection signal by transmitting and receiving radio waves every predetermined time;
Between the first position on the surface of the road below the sensor unit and the second position where the radio wave that has passed close to the rear end of the traveling vehicle in the upper part and in the traveling direction is reflected on the road. And the first distance acquired at every two different predetermined times, and the second distance from the first position to the front end in the traveling direction of the judged vehicle By calculating using the second distance, a detection length related to the length of the vehicle to be determined and a detection height related to the height of the vehicle to be determined are obtained, and based on the detection length and the detection height, And a vehicle shape feature determination unit that determines a shape feature.
(Appendix 2)
An equipment noise level that is present in the detection signal at every predetermined time and that is not caused by a vehicle including the road and the vehicle to be judged, and an intensity of a noise signal, and in the detection signal at the predetermined time The vehicle to be determined as a determination target among the vehicles by comparing the intensity of the detection signal from the vehicle with each of the road surface reflection level that is the intensity of the detection signal from the road existing in the vehicle A level determination unit to identify,
The vehicle detection device according to appendix 1, wherein the sensor unit further detects the vehicle including the vehicle to be determined as a detection signal by transmitting and receiving radio waves every predetermined time.
(Appendix 3)
The vehicle shape feature determination unit uses a plurality of the first distances and the second distances acquired using the two different predetermined times, and uses a plurality of first distances and a plurality of second distances. As a feature of the shape of the judged vehicle, the plurality of detection lengths and the plurality of detection heights corresponding to the above are acquired, the transition of the plurality of detection lengths and the transition of the plurality of detection heights are determined in order of elapsed time. Appendix 1 or Appendix 2 for determining the vehicle length of the vehicle to be determined, the vehicle height of the vehicle to be determined, the length of a portion of the vehicle to be determined, and / or the height of a portion of the vehicle to be determined The vehicle detection device described in 1.
(Appendix 4)
The vehicle shape feature determination unit is configured to determine the vehicle shape and the target of the determination target vehicle based on the intensity of the detection signal of the determination target vehicle acquired from the level determination unit and the shape characteristic of the determination target vehicle. The vehicle detection device according to any one of supplementary notes 1 to 3, which determines a vehicle type of the determination vehicle.
(Appendix 5)
The level determination unit and / or the vehicle shape feature determination unit determines the speed of the vehicle to be determined based on the first position of the vehicle to be determined at each of the plurality of predetermined times. The vehicle detection device according to claim 1.
(Appendix 6)
In determining the transition of the plurality of detection lengths and the transition of the plurality of detection heights in the order of the elapsed time, the vehicle shape feature determination unit has the plurality of detection lengths and the plurality of detection heights in the vicinity of the sensor unit. If the vehicle shape of the vehicle to be judged is a box shape, the vehicle shape of the vehicle to be judged is changed when the plurality of detection lengths and the plurality of detection heights increase or decrease. The vehicle detection device according to any one of appendices 3 to 5, which is determined not to have a box shape.
(Appendix 7)
The level determination unit uses the threshold value of the intensity of a plurality of types of vehicle reflection signals according to the size of the vehicle and the height of the vehicle, the device noise level, and the road surface reflection level to correspond to the detection signal. The vehicle detection device according to any one of appendices 2 to 6, which determines a position where a road exists, a position where a plurality of types of the vehicles exist according to a size of the vehicle and a height of the vehicle.
(Appendix 8)
Furthermore, a motor for scanning the sensor unit,
The vehicle detection device according to any one of appendices 1 to 7, further comprising: a sensor control unit that controls the motor to cause the motor to perform scanning at a period of each predetermined time.
(Appendix 9)
Furthermore, the said sensor part is plurality, The vehicle detection apparatus as described in any one of the supplementary notes 1 thru | or 7 arrange | positioned so that these sensor parts may each transmit and receive the said electromagnetic wave in the same direction.
(Appendix 10)
The vehicle detection device according to any one of appendices 1 to 9, wherein the radio wave is a millimeter wave.
(Appendix 11)
Detecting the road and the vehicle to be determined as detection signals by transmitting and receiving radio waves from the upper part of the road toward the road every predetermined time,
The first position on the surface of the road below the position where the detection signal is detected, and the radio wave that has passed close to the top of the vehicle to be judged and the rear end in the traveling direction is reflected on the road. The first distance between two positions and the second distance from the first position to the front end in the traveling direction of the vehicle to be judged are acquired at two different predetermined times. By calculating using the first distance and the second distance, a detection length related to the length of the vehicle to be determined and a detection height related to the height of the vehicle to be determined are obtained, and based on the detection length and the detection height A vehicle detection method for determining a shape characteristic of the vehicle to be determined.
(Appendix 12)
A stationary body that does not move is installed thereon, and a movable body that moves is installed above a reference plane that moves on the stationary body, receives electromagnetic waves every predetermined time, and uses the stationary body and the movable body as detection signals. A sensor unit to detect;
A first position on a surface of the reference surface vertically below the sensor unit; and a second position at which the electromagnetic wave that has passed close to the rear end of the moving body and in the traveling direction is reflected on the reference surface. A first distance between the first position and the second distance from the first position to the front end of the moving body in the traveling direction, and the first distance acquired at two different predetermined times. And the second distance are used to calculate a detection length related to the length of the moving body and a detection height related to the height of the moving body, and the shape of the moving body is determined based on the detection length and the detection height. And a determination unit that determines characteristics.

DESCRIPTION OF SYMBOLS 1 Vehicle detection apparatus 2 Sensor apparatus 3 Signal processing apparatus 5 Communication center 6 Information provision apparatus 7 Vehicle 13 Intersection 15 Road 16 Strut 21 Sensor part 23 Sensor control part 31 Signal detection part 32 Tracking part 33 Car number recognition part 73 Right turn car 77 Opposite Car 78 Rear opposite vehicle 151 Roadway 153 Traveling side roadway 157 Opposite side roadway 211 Antenna 212 Scanning motor 213 Transmission / reception switching switch 214 Antenna connection switch 215 Reception side amplifier 216 Reception side A / D converter 218 Transmission side amplifier 219 Transmission side D / A converter 311 Signal input control unit 312 Level determination unit 313 Reference level storage unit 314 Determination information storage unit 315 Vehicle length vehicle height calculation unit 316 Vehicle length vehicle height storage unit 317 Vehicle shape vehicle type determination unit 318 Communication unit 319 Vehicle shape feature determination unit BL reference Level N L Equipment noise level RL Road surface reflection signal level VS1 Small vehicle vehicle reflection signal VS2 Large vehicle vehicle reflection signal VSI1 Small vehicle vehicle reflection signal strength VSI2 Large vehicle vehicle reflection signal strength H Sensor height from road h Detection height c Detection length F Horizontal distance from the sensor part to the front edge of the vehicle body L Horizontal distance from the sensor part to the road detection position of the sensor part B Horizontal distance from the rear edge part of the upper part of the vehicle body to the road detection position of the sensor part

Claims (5)

A sensor unit that is installed above the road and detects the road and the vehicle to be determined as a detection signal by transmitting and receiving radio waves every predetermined time;
Between the first position on the surface of the road below the sensor unit and the second position where the radio wave that has passed close to the rear end of the traveling vehicle in the upper part and in the traveling direction is reflected on the road. And the first distance acquired at every two different predetermined times, and the second distance from the first position to the front end in the traveling direction of the judged vehicle By calculating using the second distance, a detection length related to the length of the vehicle to be determined and a detection height related to the height of the vehicle to be determined are obtained, and based on the detection length and the detection height, And a vehicle shape feature determination unit that determines a shape feature. An equipment noise level that is present in the detection signal at every predetermined time and that is not caused by a vehicle including the road and the vehicle to be judged, and an intensity of a noise signal, and in the detection signal at the predetermined time The vehicle to be determined as a determination target among the vehicles by comparing the intensity of the detection signal from the vehicle with each of the road surface reflection level that is the intensity of the detection signal from the road existing in the vehicle A level determination unit to identify,
The vehicle detection device according to claim 1, wherein the sensor unit further detects the vehicle including the vehicle to be determined as a detection signal by transmitting and receiving radio waves every predetermined time.   The vehicle shape feature determination unit uses a plurality of the first distances and the second distances acquired using the two different predetermined times, and uses a plurality of first distances and a plurality of second distances. As a feature of the shape of the judged vehicle, the plurality of detection lengths and the plurality of detection heights corresponding to the above are acquired, the transition of the plurality of detection lengths and the transition of the plurality of detection heights are determined in order of elapsed time. The vehicle length of the judged vehicle, the vehicle height of the judged vehicle, the length of a part of the judged vehicle, and / or the height of a part of the judged vehicle are determined. Item 3. The vehicle detection device according to Item 2.   The vehicle shape feature determination unit is configured to determine the vehicle shape and the target of the determination target vehicle based on the intensity of the detection signal of the determination target vehicle acquired from the level determination unit and the shape characteristic of the determination target vehicle. The vehicle detection device according to any one of claims 1 to 3, wherein a vehicle type of the determination vehicle is determined. Detects roads and vehicles using sensors that transmit and receive radio waves at predetermined times,
A first position where a vertical line from a position where the sensor is installed intersects the surface of the road; and a second position on the road where radio waves that have passed near the rear end of the vehicle and in the traveling direction are reflected on the road. A first distance between two positions and a second distance between the first position and the front end of the traveling direction of the vehicle for two different predetermined times, and A vehicle detection method used for calculating a length and / or a height of the vehicle.

Images