JP5625420B2 - Vehicle detection device and vehicle detection system - Google Patents

Description

  The present invention relates to an apparatus and a system for detecting a vehicle using a radar.

  In recent years, a system for ensuring safety on road traffic and providing traffic information by notifying vehicle drivers, pedestrians, etc. using road panels, delineators, in-vehicle devices, etc., on road traffic volume and traffic jam information. It is being deployed. In such a system, for example, a detection device equipped with a sensor module such as a visible camera, an infrared camera, a millimeter wave radar, or a laser radar is installed at a location where accidents occur frequently, and the analysis result of the data obtained from the sensor module To the driver. Among the detection devices, the millimeter wave radar has an advantage that an error in detection results due to a change in weather is less likely to occur.

  As a related technology, based on the reflected wave of the radio wave transmitted from the radio wave radar transmitter that transmits the radio wave toward the A section on the road, the position in the coordinate system based on the ground including the road surface in the A section Is known, and a calculation result is transmitted to the vehicle in the A section.

Japanese Patent Laid-Open No. 11-86183

  When providing information on a vehicle located in front of the lane in which the vehicle is traveling to the traveling vehicle, information for identifying the lane in which the vehicle is located is used. For example, in a system for preventing accidents such as rear-end collisions, when one lane on a two-lane road is congested, it provides traffic information to vehicles traveling in the congested lane, It is desirable not to provide traffic information to vehicles traveling in the lane. Therefore, in order to identify the lane in which the vehicle is located, one radar may be installed for each lane. However, when a radar with a sharp beam width is used, information on a plurality of lanes may be acquired by one detection device, and the lane in which the detected vehicle is traveling may not be specified. On the other hand, if a radar with a sharp beam width is used, there is a possibility that a vehicle traveling near one side of the lane cannot be detected. Although it is conceivable to use a scan type radar, the use of the scan type radar has a problem that the service life of the system is significantly shortened because the service life of the movable part for scanning is short. These problems are not limited to millimeter wave radars, but may be considered when any type of radar is used for vehicle detection.

  An object of the present invention is to determine a lane in which a vehicle detected by a radar is traveling.

  The vehicle detection device according to the embodiment is installed in each of a plurality of lanes, and detects a vehicle using information from a radar that includes a lane adjacent to the installation lane as a detection range. The vehicle detection device includes an acquisition unit, an observation radar identification unit, and a determination unit. The acquisition unit acquires, from each radar, a detection distance representing a distance between the detected object and the radar that detected the object, and a detection time of the object. The observation radar specifying unit detects the detection distance of the first object detected by the first radar and the second radar installed in the lane adjacent to the lane where the first radar is installed. The distance observation radar which observed the shortest distance among the detection distances of the two objects is specified. The determination unit determines that the vehicle is traveling in a lane in which the distance observation radar is installed when a distance between the first object and the second object is equal to or less than a first threshold value. To do.

  The lane in which the vehicle detected by the radar is traveling can be determined.

It is a figure which shows the example of the system by which the vehicle detection apparatus concerning embodiment is used. It is a figure explaining an example of composition of a vehicle detection device. It is a figure which shows an example of an installation position table. It is a figure which shows the example of the detection data table which the vehicle detection part produced | generated. It is a figure which shows an example of an input table. It is a flowchart explaining an example of operation | movement of the acquisition part at the time of producing | generating an input table. It is a figure explaining the example of the combination of the running state of a vehicle, and the observation result of a radar. It is a flowchart explaining an example of operation | movement of a control part. It is a figure which shows an example of an integrated data table. It is a figure which shows the example of the system which installed the vehicle detection apparatus and the radar. It is a figure explaining an example of the composition of the vehicle detection device concerning a 2nd embodiment. It is a figure which shows an example of an input table. It is a flowchart explaining an example of operation | movement of a control part. It is a figure explaining an example of the composition of the vehicle detection device concerning a 3rd embodiment. It is a figure explaining the example of the combination of the running state of a vehicle, and the observation result of a radar. It is a figure explaining an example of the determination method of an electric power difference threshold value. It is a flowchart explaining an example of operation | movement of a control part. It is a figure explaining an example of the composition of the vehicle detection device concerning a 4th embodiment. It is a flowchart explaining an example of operation | movement of a control part. It is a flowchart explaining an example of operation | movement when combining 1st-3rd embodiment. It is a figure which shows the example of the detection data table of time t1, and an input table. It is a figure explaining an example of the comparison process of the target object in an integrated data table, and the target object in an input table. It is a flowchart explaining an example of the comparison process of the target object in an integrated data table, and the target object in an input table. It is a figure which shows an example of the integrated data table after deleting the data regarding the vehicle in which the corresponding target object is not detected. It is a flowchart explaining an example of the production | generation method of an integrated data table. It is a flowchart explaining an example of the production | generation method of an integrated data table. It is a figure which shows the example of an integrated data table and an input table. It is a flowchart explaining an example of a determination process. It is a flowchart explaining an example of a determination process. It is a flowchart explaining an example of an integration process. It is a flowchart explaining an example of the method of new registration. It is a flowchart explaining an example of the determination between integrated data.

Hereinafter, embodiments will be described in detail with reference to the drawings.
FIG. 1 is a diagram illustrating an example of a system in which the vehicle detection device according to the embodiment is used. In the following description, it is assumed that the vehicle detection device 20 detects a vehicle traveling in each of the lanes 1 to 3. In the following description, a lane number is assigned to each lane so that the number becomes smaller as the left-hand side in the traveling direction of the vehicle. However, the lane number can be arbitrarily determined.

  Radars 11 to 13 are radars that detect objects located in lanes 1 to 3. Here, the radar used does not have position resolution in the direction perpendicular to the traveling direction of the lane, and can be any radar or sensor whose detection range extends over a plurality of lanes. In the following description, the radars 11 to 13 are assumed to be millimeter wave radars. In the following description, it is assumed that any radar detects an object located 42 to 122 meters from the installation position. Further, in any radar, in the range of 60 to 120 meters from the installation position, the lane adjacent to the lane where the radar is installed is included in the coverage area 4 (4a to 4c) of the radar. Here, it is assumed that the covered area 4 a is the radar 11, the covered area 4 b is the radar 12, and the covered area 4 c is the covered area of the radar 13. In the following description, it is assumed that the installation positions of the radars 11 to 13 with respect to the traveling direction of the lane are substantially the same. The radars 11 to 13 observe the target object included in the covered area 4 and output the observation result to the vehicle detection device 20 at regular time intervals such as every 100 ms. Note that the time interval at which the radars 11 to 13 observe the object is arbitrary.

  In the following description, an object detected by each of the radars 11 to 13 may be described as a target. Therefore, for example, when the radar 11 detects the vehicle A, it may be described as detecting the object A corresponding to the vehicle A. Further, when the radar 11 detects the vehicle C traveling in the lane 2, the radar 11 is expressed as detecting the object Cd corresponding to the vehicle C. Here, a vehicle detected by a radar provided in a lane other than the driving lane is indicated by “Cd” by adding a lowercase letter d to the alphabet indicating the vehicle. On the other hand, an actual vehicle traveling in one of the lanes 1 to 3 or a vehicle in which the traveling lane is determined is represented as a vehicle, such as the vehicle A.

  FIG. 1A shows an example of the traveling state of the vehicle at time t0. In the following description, it is assumed that the radars 11 to 13 observe the object at time t1 after time t0. FIG. 1B shows an example of the traveling state of the vehicle at time t1. At time t0, five vehicles A to E are traveling, and four vehicles A to D are within the detection range of radars 11 to 13. Here, Ad, Bd, and Cd each indicate an object detected by a radar installed in a lane adjacent to a traveling lane. Next, it is assumed that the vehicle A exits the detection range of the radars 11 to 13 and the vehicle E enters the detection range of the radars 11 to 13 at time t1. In the following description, first, the operation of the vehicle detection device 20 will be described by taking as an example the case of processing information detected by the radars 11 to 13 at time t0, and the case of processing information at time t1 will be described later.

<First Embodiment>
FIG. 2 is a diagram illustrating an example of the configuration of the vehicle detection device 20. The vehicle detection device 20 includes a communication unit 31 (31a to 31c), a memory 32, a detection unit 33 (33a to 33c), a control unit 40, and a transmission unit 34. The control unit 40 includes an acquisition unit 41, an integration unit 42, a target extraction unit 51, a distance observation radar specifying unit 53, a determination unit 56, and an output unit 43.

  The vehicle detection device 20 is connected to the radars 11 to 13 and acquires data measured by these radars. The radars 11 to 13 notify the vehicle detection device 20 of the position of the object included in each covered area. When the vehicle detection device 20 acquires data from the radars 11 to 13, the vehicle detection device 20 refers to the installation position table and inputs the acquired data to the communication units 31 a to 31 c. In the following description, it is assumed that the installation position table is stored in the memory 32. In addition to the installation position table, the memory 32 can store any table or data used in the vehicle detection device 20 such as an input table described later.

  FIG. 3 is a diagram illustrating an example of the installation position table. In the example of FIG. 3, the port number, radar number, and lane number of the vehicle detection device 20 are recorded in association with each other. Note that the table shown in FIG. 3 is an example of an installation position table, and information included in the installation position table can be arbitrarily changed according to mounting. When data is input to the vehicle detection device 20, the vehicle detection device 20 confirms the number of the port where the data is input. When the port number is recorded in the installation position table, the vehicle detection device 20 determines that the input is from the radar, and outputs the data input to the communication unit 31. For example, when data is input from a port 38811, the vehicle detection device 20 outputs the input data to the communication unit 31a. Similarly, the vehicle detection device 20 outputs data input from the 38812 port to the communication unit 31b, and outputs data input from the 38813 port to the communication unit 31c.

  The communication units 31a to 31c output data input from the vehicle detection device 20 to the detection units 33a to 33c. The detection unit 33 receives data such as the moving speed of the object from the communication units 31a to 31c, and performs noise removal, background processing, tracking processing, and the like on the received data. The detection unit 33 generates a detection data table based on the data measured by the radar 11 by attaching an object ID based on the moving speed of the object in the tracking process. An example of the detection data table and a usage method will be described later. In addition, the detection parts 33a-33c can use arbitrary methods for processes, such as noise removal, a background process, and a tracking process.

  The acquisition unit 41 acquires a detection data table from the detection units 33a to 33c. Note that the detection units 33a to 33c can store the detection data table in the memory 32 and notify the acquisition unit 41 accordingly. The object extraction unit 51 extracts a combination of objects that may be one vehicle among a plurality of objects detected by the radar. As described above, in the range of 60 to 120 meters from the installation position, the vehicle that travels in the lane adjacent to the lane where the radar is installed is included in the coverage area 4 of the radar. Therefore, the target extraction unit 51 refers to the installation position table, identifies the first lane and the second lane adjacent to the first lane, and detects the object detected in the first and second lanes. Of these, it is determined that there is a possibility that the two objects whose detection positions are close are one vehicle. The target extraction unit 51 outputs the extracted combination to the distance observation radar specifying unit 53 and the determination unit 56. The operation of the target extraction unit 51 will be described in detail later.

  The distance observation radar specifying unit 53 specifies a radar that has detected any of the objects extracted by the target extracting unit 51 at a position closest to the radar by comparing the history of detection results of the respective radars. The distance observation radar specifying unit 53 notifies the determination unit 56 of the specified result. The operation of the distance observation radar specifying unit 53 will be described in detail later.

  The detection start position of a vehicle traveling in a lane where a radar is installed is closer to the radar than the detection start position of a vehicle traveling in a lane adjacent to the lane where the radar is installed become. In the example of FIG. 1A, the radar 11 can detect a vehicle traveling on the lane 1 at 42 meters from the radar 11, but a vehicle traveling on the lane 2 can be detected only after the radar 11 is 61.5 meters or later. Therefore, the determination unit 56 determines that the vehicle is traveling in the lane in which the radar specified by the distance observation radar specifying unit 53 is installed. The combination extracted by the target extraction unit 51 will be described in detail later on the operation of the control unit 40 including the determination unit 56.

  The determination unit 56 outputs the determination result to the integration unit 42. The integration unit 42 integrates the object data according to the determination result of the determination unit 56. The integration unit 42 outputs the integrated data to the output unit 43. The output unit 43 outputs the input result to the transmission unit 34. The transmission unit 34 transmits the result input from the output unit 43 to the host device 60 that communicates with the vehicle detection device 20. The host device 60 notifies the subsequent vehicle of the running state of the vehicle using an optical beacon or the like.

  By using such a vehicle detection device 20, even when the radar coverage 4 extends over a plurality of lanes and the radar does not have a resolution in a direction perpendicular to the traveling direction of the lanes, detection is performed by the radar. The lane in which the target object travels can be specified.

Hereinafter, the process of the control unit 40 in the case of FIG. 1A will be described in detail.
FIG. 4 is a diagram illustrating an example of the detection data table generated by the detection units 33a to 33c. The detection unit 33a generates the detection data table in FIG. 4A using the data input from the communication unit 31a. Similarly, the detection parts 33b and 33c process the data input from the communication parts 31b and 31c similarly to the detection part 33a, and produce | generate the detection data table shown in FIG.4 (b) and FIG.4 (c). The detection data table generated by the detection units 33 a to 33 c is output to the acquisition unit 41.

  The detection data table shown in FIG. 4 includes the observation time, the identifier of the object, the distance between the object detection position and the radar that detected the object, the moving speed of the object, and the reception received by the radar from the object. Includes electricity. In the following description, a distance between a position where an object is detected and a radar that detects the object may be referred to as a “detection distance”. In the following description, each piece of data for each object included in the detection data table may be referred to as “detection data”.

  As shown in FIG. 1A, since the detection range of the radar 11 includes vehicles A and B, the radar 11 detects the object A and the object B. Further, the lane 1 and the lane 2 are included in the covered area 4 a of the radar 11 after the point 60 m from the installation position of the radar 11. Accordingly, the radar 11 also detects the object Cd corresponding to the vehicle C traveling in the lane 2. In addition, the alphabet which shows the target object described in the left side of the line contained in the table of Fig.4 (a)-FIG.4 (c) is shown in order to assist an understanding, and in a detection data table, Not included.

  The vehicle C is included in the detection range of the radar 12. Further, the covered area 4b of the radar 12 includes the lane 1 and the lane 3 after 60 m from the installation position. Therefore, the radar 12 also detects the object Ad and the object Bd corresponding to the vehicles A and B traveling in the lane 1. Since the radar 12 includes the vehicle D traveling in the lane 3 in the covered area 4b, the radar 12 also receives the reflected wave reflected from the vehicle D. However, in the example of FIG. 1A, since the vehicle D and the vehicle C are running in parallel, the radar 12 cannot distinguish between the vehicle C and the vehicle D, and the vehicle C and the vehicle D are combined 1 Detects one object. Therefore, when vehicles are running side by side, such as vehicle C and vehicle D, the radar detects a vehicle traveling in the lane in which the radar is installed. For example, the data in the third row in FIG. 4B is data of the object C corresponding to the vehicle C. Thus, based on the observation result of the radar 12, as shown in FIG. 4B, detection data relating to the object C, the object Ad, and the object Bd is generated.

  The radar 13 includes the vehicle D traveling in the lane 3 in the covered area 4c. Since the vehicle C and the vehicle D are running side by side, the radar 13 detects the object D corresponding to the vehicle D as described above. 4A to 4C are examples of the detection data table, and the information included in the detection data table can be arbitrarily changed according to the implementation. For example, the detection units 33 a to 33 c may output a detection data table that does not include the moving speed and reception intensity of the target object to the acquisition unit 41.

  Next, the operation of the acquisition unit 41 will be described. The acquisition unit 41 generates an input table at time t0 based on the detection data table acquired from the detection units 33a to 33c. FIG. 5 is a diagram illustrating an example of the input table. In this example, the data in the first to third lines of the input table correspond to the detection data table of FIG. The data in the 4th to 6th lines of the input table corresponds to the detection data table in FIG. 4B, and the data in the 7th line of the input table corresponds to the detection data table in FIG. In addition, in order to make it easy to see the correspondence between the data in the input table and the detection data, an alphabet indicating the object is described on the left side of the input table, but this alphabet is not included in the input table.

In addition to the information contained in the detection data table, the input table includes the time at which each radar was measured, the detection distance when each object was first detected, the weighted average value of the received power, the marker including. A method of using the marker will be described later. The acquisition unit 41 calculates the weighted average value of the received power, for example, according to the following formula, but may be calculated using another calculation formula.
Received power weighted average = previous received power value x α + current received power value × (1-α)
Here, α is a coefficient less than 1.

  FIG. 5 is an example of an input table, and the acquisition unit 41 can generate an input table that does not include, for example, a marker or a weighted average value of received power. The acquisition unit 41 records the generated input table in the memory 32.

  FIG. 6 is a flowchart illustrating an example of the operation of the acquisition unit 41 when generating an input table. In the flowchart of FIG. 6, the acquisition unit 41 generates a new input table using the input table generated in the previous process. For example, if the time when the object is observed before time t0 is tz, the acquisition unit 41 uses the input table generated based on the observation result at time tz to input using the observation result at time t0. Generate a table. The acquisition unit 41 uses variables n and x and constants N and X. Any of n, x, N, and X can be any positive integer. n is a variable for counting the number of radars that have processed the detection data table, and x is a variable for counting the number of detection data compared with the data in the input table. N represents the total number of radars that output data to the vehicle detection device 20, and X represents the total number of detection data included in the detection data table to be processed. Here, the operation when the input table at time t0 shown in FIG. 5 is generated using the input table at time tz will be described.

  The acquisition unit 41 sets the value of n to 1 (step S1a). The acquisition unit 41 sets the value of x to 1 (step S1b). Next, the acquisition unit 41 confirms whether the identifier of the x-th object observed by the radar exists in the input table at time tz for each detection data included in the n-th detection data table. (Step S2). When the identifier of the xth object observed by the nth radar exists in the input table at time tz, the acquisition unit 41 updates the input information of the object to the information in the detection data table at time t0. (Step S3). On the other hand, when the identifier of the xth object observed by the nth radar does not exist in the input table at time tz, the acquisition unit 41 uses the information in the detection data table for the object as the input table at time t0. (Step S4). When the process of step S3 or S4 ends, the acquisition unit 41 increments the value of x by 1 and compares it with X (step S5). When x is a value equal to or smaller than X, the acquisition unit 41 repeats the processes of steps S2 to S4. When x is larger than X, all the objects included in the detection data table to be processed are compared with the input table. Therefore, the acquisition unit 41 increments the value of n by 1 and compares it with N (step S6). When n is a value equal to or smaller than N, the acquisition unit 41 determines that there is a detection data table that has not yet been processed, and repeats the processing of steps S2 to S5 for the next detection data table. If n is greater than N, it is determined that all the detection data tables have been processed. Therefore, the acquisition unit 41 deletes data about an object whose information has not been updated or added among the objects included in the input table (step S7).

  When the input table is generated, the target extraction unit 51 extracts a target for determination processing from the input table. The object extraction unit 51 is detected by an object detected by the first radar installed in the first lane and by a second radar installed in the second lane adjacent to the first lane. Compare the detection distance for the object. When the distance (D1) of the first object from the first radar and the distance (D2) of the second object from the second radar are short, the first object and the second object are Since there is a possibility that they are the same vehicle, the target extraction unit 51 sets the first and second objects as processing targets. Therefore, in order to extract the target of the determination process, the target extraction unit 51 compares, for example, the absolute value of the value of D1-D2 with the first threshold value. When the absolute value of the value of D1-D2 is smaller than the first threshold value, the target extraction unit 51 sets the first target object and the second target object as objects of the determination process. Here, it means that the first object and the second object to be subjected to the determination process are two objects that may be the same vehicle. Accordingly, the first threshold is detected based on the object detected based on the reflected wave from the front or middle of the large vehicle, such as a value of about 3.5 m, and the reflected wave from the rear of the large vehicle. The value is large enough to determine that there is a possibility that the object is a single vehicle.

  Hereinafter, an operation example when the object extraction unit 51 extracts the object A and the object Ad from the input table will be described. Note that the target extraction unit 51 may perform (2) before (1) in the following procedure.

(1) The object extraction unit 51 refers to the input table and recognizes that the detection distance of the object (object A) with the object ID = 1 detected by the radar 11 is 106.5 m.
(2) The target extraction unit 51 refers to the installation position table and confirms that the radar 11 is installed in the lane 1. Further, the target extraction unit 51 also recognizes that the radar installed in the lane 2 adjacent to the lane 1 is the radar 12.

  (3) In the input table, the detection distance of the object obtained based on the measurement data from the radar 12 is compared with the detection distance of the object with the object ID = 1 detected by the radar 11. The detection distance of the object (object Ad) having the object ID = 1 detected by the radar 12 is 106.8 m. Accordingly, the difference in the detection distance between the object with the object ID = 1 detected by the radar 11 (object A) and the object with the object ID = 1 detected by the radar 12 (object Ad) is 0.3 m. Therefore, there is a possibility that the object A and the object Ad are the same vehicle. Therefore, the target extraction unit 51 sets the target object A and the target object Ad as processing targets.

  The object extraction unit 51 performs the same process on other objects. As a result, the target object B and the target object Bd may also be a single vehicle, and thus are processed. Further, the combination of the object Cb and the object C and the combination of the object C and the object D are also subject to processing. Here, for the sake of easy understanding, the object B and the like are described, but the object extraction unit 51 uses the identifier of the radar that detected the object and the object ID for each object. Are identified. The target extraction unit 51 notifies the distance observation radar specifying unit 53 and the determination unit 56 of the processing target.

  The distance observation radar specifying unit 53 refers to the input table and acquires the initial value of the detection distance for each target object specified as the processing target. The acquired initial value of the detection distance is compared, and the radar in which the smaller value is detected is specified. For example, the initial value of the detection distance of the object (object A) with the object ID = 1 detected by the radar 11 is 41 meters. On the other hand, the initial value of the detection distance of the object (object Ad) having the object ID = 1 detected by the radar 12 is 62.2 meters. Since the initial value of the detection distance observed by the radar 11 is smaller than that detected by the radar 12, the distance observation radar specifying unit 53 specifies the radar 11. In the following description, among the radars that have observed the object to be processed, the radar having the smaller initial detection distance may be referred to as “distance observation radar”. It can also be said that the distance observation radar is a radar that observes the shortest distance among the detection distances of the target object to be processed. The distance observation radar specifying unit 53 notifies the determination unit 56 of the distance observation radar.

  FIG. 7 is a diagram for explaining an example of a combination of the traveling state of the vehicle and the observation result of the radar. The operation of the determination unit 56 will be described with reference to FIG. FIG. 7A is a diagram showing the traveling state of vehicle A and lanes 1 and 2 in FIG. FIG. 7B shows the change over time of the detection distance of the object A (object ID = 1) in the radar 11, and FIG. 7C shows the object Ad (object ID = 1) in the radar 12. It represents the change over time of the detection distance. The point at which the first radar installed in the first lane in which the vehicle is traveling begins to detect the vehicle is the second radar installed in the second lane adjacent to the first lane. It is a point closer to the first and second radars than the point where the vehicle is detected. For example, as shown in FIGS. 7A to 7C, the position where the radar 11 starts to detect the vehicle A traveling in the lane 1 starts to detect the vehicle in which the radar 12 travels in the lane 1. The position is closer to the radar 11 than the position.

  When notified of the distance observation radar from the distance observation radar specifying unit 53, the determination unit 56 determines that the vehicle is located in the lane in which the distance observation radar is installed. In addition, the determination unit 56 appropriately refers to the installation position table and confirms the lane in which the distance observation radar is installed. For example, it is assumed that the determination unit 56 is notified that the distance observation radar is the radar 11 for the object A and the object Ad. Then, the determination unit 56 determines that the object A and the object Ad correspond to one vehicle traveling on the lane 1 based on the installation position table.

  Further, by comparing the distance difference threshold value with the difference between the initial values of the detection distances of the distance observation radar and the other radar, the determination unit 56 can determine whether two vehicles are running side by side or one vehicle. It is also possible to detect the vehicle corresponding to both cases where the vehicle is traveling. In this case, the distance observation radar specifying unit 53 obtains a difference between the initial values of the detection distances of the distance observation radar and the other radar, and outputs the difference to the determination unit 56 together with the identifier of the distance observation radar. For example, in the example of the object A and the object Ad, the determination unit 56 is notified that 62.2−41 = 21.2 meters is the difference between the initial values of the detection distances.

  Next, an example of how to obtain the distance difference threshold will be described. For example, the radar 12 installed in the lane 2 adjacent to the lane 1 in which the vehicle A is traveling cannot detect the vehicle A until the lane 1 enters the covered area 4b. Therefore, the distance difference threshold value is the same length as the distance from the position at which a certain vehicle starts to be detected by the radar in the traveling lane to the position where the radar coverage 4 installed in the adjacent vehicle overlaps. Is set. However, the position detected by the radar may vary depending on the size of the traveling vehicle. For example, when only one large vehicle such as a bus or truck travels in one lane, it is easy to receive reflected waves even with a radar installed in the lane where the large vehicle is not traveling. For this reason, a large vehicle is detected from a position closer to the radar than a normal vehicle or a small vehicle. On the other hand, motorcycles and small cars are difficult to detect because of their small surface area. Therefore, for example, when a two-wheeled vehicle is traveling on the lane 1, the radar 12 may not detect an object corresponding to the two-wheeled vehicle even if the two-wheeled vehicle reaches a point where the covered area 4 b and the lane 1 overlap. However, when the truck is traveling on the lane 1, the radar 12 may detect an object corresponding to the truck as soon as the truck reaches the point where the covered area 4 b and the lane 1 overlap. Accordingly, the distance difference threshold is set so that the lane in which the vehicle is traveling can be correctly identified even when the vehicle detected at a position close to the radar is traveling.

  The determination unit 56 confirms whether the difference in the initial value of the detection distance of the object is a value equal to or greater than the distance difference threshold value. When the difference between the initial values of the detection distances is equal to or greater than the distance difference threshold, the determination unit 56 determines that one vehicle is traveling in the lane where the distance observation radar is installed. For example, it is assumed that the distance difference threshold is 18 meters. The difference in the initial value of the detection distance between the object A and the object Ad is 21.2 meters as described above, and the distance observation radar is the radar 11 installed in the lane 1. Therefore, the determination unit 56 determines that the object A and the object Ad are detection results of one vehicle traveling in the lane 1. Here, examples of the object A and the object Ad have been described, but the object B and the object Bd are processed in the same manner.

  FIG. 7 (d) is a diagram showing the traveling state of vehicle C and lanes 1 and 2 in FIG. FIG. 7E shows the change over time of the detection distance of the object Cd (object ID = 3) in the radar 11, and FIG. 7F shows the object C (object ID = 3) in the radar 12. It represents the change over time of the detection distance. Based on the input table, the distance observation radar specifying unit 53 obtains a distance observation radar and the like. In the object C and the object Cd, the distance observation radar is the radar 12, and the difference between the initial values of the detection distances is 61.6−42.3 = 19.3 meters. Since the difference between the initial detection distance values is equal to or greater than the distance difference threshold, the determination unit 56 determines that the object C and the object Cd are detection results of one vehicle traveling in the lane 2. To do.

  When the determination between the object detected by the radar 11 and the object detected by the radar 12 is completed, the determination unit 56 next detects the object detected by the radar 12 and the object detected by the radar 13. Judge between. Here, it is determined by the determination between the object Cd and the object C that the object Cd and the object C are one vehicle traveling in the lane 2. However, since the object C and the object D may correspond to the same vehicle, the determination unit 56 determines between the object C and the object D.

  FIG. 7 (g) is a diagram showing the traveling state and lanes 2 and 3 of the vehicles C and D in FIG. FIG. 7H shows the change over time of the detection distance of the object C (object ID = 3) by the radar 12, and FIG. 7I is the detection distance of the object D (object ID = 1) by the radar 13. Represents the change over time. In the case shown in FIG. 7G, the radar 12 starts to detect the object C when the vehicle C enters the covered area 4 b of the radar 12 in the lane 2. Accordingly, the radar 12 detects the object C from a position 42.3 meters from the radar 12. On the other hand, the radar 13 starts detecting the object D when the vehicle D enters the covered area 4 c of the radar 13 in the lane 3, and thus detects the object D from a position 41.3 meters from the radar 13. . That is, when vehicles are present in both lane 2 and lane 3, both radar 12 and radar 13 are installed from the time when the vehicle enters approximately the same distance from the radar. Detects a vehicle traveling in a lane. Based on the input table, the distance observation radar specifying unit 53 is that the distance observation radar is the radar 12, and the difference between the initial values of the detection distances of the object C and the object D is 42.3-41.3 = 1. Ask for 0 meters. The determination unit 56 determines that the vehicle is located in either the lane 1 or the lane 2 because the difference between the initial values of the detection distances of the object C and the object D is smaller than the distance difference threshold. The determination unit 56 outputs the determination result to the integration unit 42.

  As described above, the vehicle detection device 20 uses the distance between the position where the object starts to be detected by the radar installed in each of the two adjacent lanes and the radar to determine the lane in which the vehicle is traveling. Identify. Referring to FIG. 7, the case where the vehicle is traveling in one lane and the case where two vehicles are traveling in two adjacent lanes have been described. The same processing can be performed when the vehicle travels across two lanes. For example, when the vehicle F travels across both the lane 1 and the lane 2, both the radar 11 and the radar 12 detect an object corresponding to the vehicle F when the vehicle F enters the covered areas 4a and 4b. To do. Therefore, the initial value of the detection distance is not significantly different between the radar 11 and the radar 12. Therefore, the determination unit 56 determines that the vehicle is located in both the lane 1 and the lane 2 because the difference in the initial value of the detection distance is smaller than the distance difference threshold.

  FIG. 8 is a flowchart for explaining an example of the operation of the control unit 40. The object extraction unit 51 confirms whether the two objects recorded in the input table have been detected by the radar installed in the adjacent lane (step S11). When the distance between two objects detected by radars installed in adjacent lanes is smaller than the first threshold, the object extracting unit 51 corresponds to the same vehicle. It is determined that there is a possibility (step S12). Next, the determination unit 56 confirms whether or not the difference between the initial values of the detection distances of the two objects is greater than or equal to the distance difference threshold value (step S13). When the difference between the initial values of the detection distances is equal to or greater than the distance difference threshold, the determination unit 56 determines that the two objects are both traveling in a lane in which the radar with the smaller initial detection distance is installed. Is detected (step S14). On the other hand, when the difference between the initial values of the detection distances is smaller than the distance difference threshold, the determination unit 56 determines that two vehicles are running side by side and each radar has detected a vehicle traveling in the installed lane. (Step S15). In addition, when the distance between two objects detected by radars installed in adjacent lanes is equal to or greater than the first threshold, the object extraction unit 51 corresponds to each of the two objects. It is determined that it is in progress (step S16).

  FIG. 9 is a diagram illustrating an example of the integrated data table. The integrated data table may be changed depending on the implementation, such as a table that does not include the received power value and the merge flag. The merge flag will be described later.

  The integration unit 42 integrates the data of the objects determined to have detected the same vehicle and generates an integrated data table. The vehicle ID in the integrated data table is an identifier assigned to a vehicle that is considered to be traveling by the determination of the determination unit 56. In the object ID column, the object ID of the object used for vehicle detection is recorded. Further, the integration unit 42 records an invalid value such as “−1” in the column of the target object ID for the radar that has not detected the target object used for vehicle detection. For example, vehicle ID = 1 is data corresponding to vehicle A. The objects corresponding to the vehicle A are the object ID = 1 (object A) detected by the radar 11 and the object ID = 1 (object Ad) detected by the radar 12. Therefore, the integration unit 42 records “1” that is the object ID of the object A in the field of the object ID of the radar 11 with the vehicle ID = 1. Similarly, the integration unit 42 records “1” that is the object ID of the object Ad in the field of the object ID in the radar 12. Since the radar 13 has not detected the object corresponding to the vehicle A, the integration unit 42 records “−1” in the object ID column of the radar 13. The integration unit 42 reads the lane number, the detection distance, the speed, the received power, and the like of the object detected by the radar installed in the lane determined to be traveling from the input table and corresponds to the vehicle ID. And record it in the integrated data table. For example, based on the object A and the object Ad, it is determined that the vehicle (vehicle A) with the vehicle ID = 1 is traveling in the lane 1. Therefore, the integration unit 42 reads the lane number, detection distance, speed, received power, and the like for the object A from the input table and records them in the integrated data table. Similarly, the integration unit 42 records data of vehicle ID = 2 based on the determination result of the object B and the object Bd.

  For the vehicle C, determination using data of the object Cd and the object C and determination using data of the object C and the object D are performed. In the determination between the object Cd and the object C, a determination result that the object Cd and the object C correspond to one vehicle traveling on the lane 2 is obtained. On the other hand, in the determination between the object C and the object D, a result is obtained that one vehicle is running along each of the lanes 2 and 3. Therefore, the integration unit 42 determines that there is a vehicle traveling on the lane 2 and records the data of the object Cd and the object C as vehicle ID = 3 in the integrated data table. Here, since it is determined that the vehicle ID = 3 is a vehicle traveling on the lane 2, the integration unit 42 records the detection distance and the like for the object C in the integrated data table. Furthermore, since it is determined that the vehicle corresponding to the object D is traveling in the lane 3, the integration unit 42 records data of vehicle ID = 4 based on the data of the object D.

  The integration unit 42 outputs the generated integrated data table to the output unit 43. The output unit 43 generates an output data table in which the object ID and the merge flag in the radars 11 to 13 are deleted from the integrated data table, and outputs the output data table to the transmission unit 34. The transmission unit 34 transmits the output data to the higher-level device 60.

  FIG. 10 is a diagram illustrating an example of a system in which the vehicle detection device 20, the radar 11, and the radar 12 are installed. In the example of FIG. 10, the vehicle detection device 20 is installed on a two-lane road on one side. When the output result is received by the host device 60 from the transmission unit 34, the host device 60 notifies the optical beacon 61 that three vehicles are traveling in the same lane ahead of the vehicle G. In the example of FIG. 10, the traveling vehicle information is notified to the vehicle G by the optical beacon 61, but the host device 60 uses the arbitrary device used for road-to-vehicle communication such as a radio wave beacon to the vehicle G. Information can be notified. In addition, communication between the transmission unit 34 and the higher-level device 60 and communication between the higher-level device 60 and the optical beacon 61 or the like can be communication via a line or wireless communication.

  Thus, according to the vehicle detection device 20 according to the first embodiment, each lane is detected using a radar that includes adjacent lanes in the coverage area and does not have resolution in the vehicle traveling direction and the vertical direction. A traveling vehicle can be detected. Since such a radar is not a scanning radar, it does not have a movable part that changes the scanning direction, and therefore has excellent durability. Therefore, it can be said that the system using the vehicle detection device 20 is also excellent in durability. Further, when the vehicle detection device 20 is used, a vehicle detection system can be made without using an expensive radar such as an electronic scan type radar. Furthermore, the problem that it may occur when a vehicle or a two-wheeled vehicle traveling in a position from an adjacent lane cannot be detected when a radar having a thin beam shape is used is also solved.

<Second Embodiment>
FIG. 11 is a diagram illustrating an example of the configuration of the vehicle detection device 70 according to the second embodiment. The vehicle detection device 70 includes a communication unit 31 (31a to 31c), a memory 32, a detection unit 33 (33a to 33c), a transmission unit 34, and a control unit 71. The control unit 71 includes an acquisition unit 44, an integration unit 42, an output unit 43, a target extraction unit 51, a time observation radar specifying unit 54, and a determination unit 56. The operations of the communication unit 31, the memory 32, the detection unit 33, the transmission unit 34, the integration unit 42, the output unit 43, and the target extraction unit 51 are the same as those in the first embodiment.

  The acquisition unit 44 generates the input table shown in FIG. 12 in the same manner as the acquisition unit 41. In FIG. 12, in addition to the information included in the input table of FIG. 5, the detection start time of each object is recorded. For example, it is assumed that the vehicle detection device 70 starts detection of a vehicle from t0 and detects until t10. The object A is first detected by the radar 11 at time t4. On the other hand, it is assumed that the object Ad has started to be detected by the radar 12 at time t6.

  The time observation radar specifying unit 54 acquires the detection start time for each target specified as the processing target, and specifies the radar with the earlier detection start time. In the following description, the radar with the earlier detection start time may be referred to as “time observation radar”. Here, since the time observation radar specifying unit 54 is a radar that observes the earliest time among the detection start times, it can be said that it is a radar that observes the earliest detection time among the detection times of the target object. When the target object to be processed is notified from the target extraction unit 51, the time observation radar specifying unit 54 refers to the input table shown in FIG.

  The operation of the determination unit 56 provided in the vehicle detection device 70 will be described with reference to FIGS. The time at which the first radar installed in the first lane in which the vehicle is traveling begins to detect the vehicle is the second radar installed in the second lane adjacent to the first lane. It is earlier than the time when the vehicle is detected. Therefore, the determination unit 56 determines that the lane in which the vehicle is traveling is a lane in which the time observation radar is installed. For example, the time (t4) when the radar 11 provided in the lane 1 in which the vehicle A is traveling detects the object A is the time (t6) when the radar 12 provided in the lane 2 detects the object Ad. Too early. Therefore, the time observation radar is the radar 11. The determination unit 56 determines that the object A and the object Ad correspond to a vehicle traveling in the lane 1 because the radar 11 is installed in the lane 1.

  The determination unit 56 can further use a time threshold value in order to detect a case where two vehicles travel side by side in adjacent lanes as shown in FIGS. 7 (g) to 7 (i). The time threshold is set, for example, from the time when the first radar installed in the first lane in which the vehicle is traveling starts to detect the vehicle to the second lane adjacent to the first lane. It can be the time to enter the coverage area of the installed second radar. Further, as described above, since the ease of detection differs depending on the vehicle type, for example, the time from when the truck traveling in the lane 1 enters the area covered by the radar 11 until it is detected by the radar 12 is the lane 1 It may be shorter than the time from when the traveling motorcycle enters the area covered by the radar 11 until it is detected by the radar 12. Therefore, the time threshold is set to a value that can identify the lane in which the vehicle is traveling, regardless of the type of the vehicle that is traveling. Here, it is assumed that the time threshold is t1.

  When the time threshold value is used for determination by the determination unit 56, the time observation radar specifying unit 54 notifies the determination unit 56 of the difference in detection start time together with the identifier of the time observation radar. For example, in the combination of the object A and the object Ad, the difference in detection start time is t2. On the other hand, in the combination of the object C and the object D, the difference in detection start time is zero.

  When determining, the determination unit 56 compares the difference in detection start time with a time threshold value. When the difference between the detection start times is smaller than the time threshold, it is determined that the two objects correspond to each of the two vehicles running side by side. For example, since the difference between the detection start times of the object C and the object D is 0, it is determined that two vehicles are running side by side. On the other hand, when the difference between the detection start times is equal to or greater than the time threshold, the determination unit 56 determines that the two objects correspond to one vehicle traveling in the lane in which the time observation radar is installed. For example, since the difference in detection start time between the object A and the object Ad is 2t, it is determined that the object A and the object Ad correspond to one vehicle traveling in the lane where the time observation radar is installed. To do. Since the time observation radar is the radar 11, the determination unit 56 refers to the installation position table and associates the object A and the object Ad with the vehicle traveling in the lane 1.

  FIG. 13 is a flowchart for explaining an example of the operation of the control unit 71. Steps S21, S22, and S26 are the same as steps S11, S12, and S16 described with reference to FIG. In step S23, the determination unit 56 confirms whether the difference between the detection start times of the two objects is equal to or greater than the time threshold value. When the difference between the detection start times of the two objects is equal to or greater than the time threshold value, the determination unit 56 determines that the two objects are both traveling in the lane in which the radar with the earlier detection start time is installed. It is determined that the detection result is (step S24). On the other hand, when the difference between the detection start times of the two objects is smaller than the time threshold, the determination unit 56 detects that the two vehicles are running side by side and each radar detects a vehicle traveling in the installed lane. Judgment is made (step S25).

  As described above, according to the second embodiment, the lane in which the vehicle is traveling can be specified by using the information of the time when the detection of the object is started.

<Third Embodiment>
FIG. 14 is a diagram illustrating an example of the configuration of the vehicle detection device 80 according to the third embodiment. The vehicle detection device 80 includes a communication unit 31 (31a to 31c), a memory 32, a detection unit 33 (33a to 33c), a transmission unit 34, and a control unit 81. The control unit 81 includes an acquisition unit 41, an integration unit 42, an output unit 43, a target extraction unit 51, a reception radar specifying unit 55, and a determination unit 56. The operations of the communication unit 31, the memory 32, the detection unit 33, the transmission unit 34, the acquisition unit 41, the integration unit 42, the output unit 43, and the target extraction unit 51 are the same as those in the first embodiment.

  The reception radar specifying unit 55 acquires a received power value for each target specified as a processing target, and specifies a radar having a larger received power value. In the following description, a radar having a larger reception power value may be referred to as “reception radar”. When an object to be processed is notified from the target extraction unit 51, the reception radar specifying unit 55 refers to an input table as shown in FIG. 5 and acquires a weighted average value of received power. The weighted average value is used to reduce the influence of noise. In an environment where the noise is extremely small, the reception radar specifying unit 55 may specify the reception radar based on the reception power value. The reception radar specifying unit 55 notifies the determination unit 56 of the reception radar.

  When receiving the notification from the reception radar specifying unit 55, the determination unit 56 refers to the installation position table and confirms the lane in which the reception radar is installed. For example, it is assumed that the determination unit 56 is notified that the reception radar is the radar 12. Then, the determination unit 56 can determine that the object corresponds to one vehicle traveling in the lane 2 based on the installation position table.

  The received power value becomes larger as the reflected wave from the vehicle is stronger. In a vehicle in which the vehicle is not traveling, the power value reflected from the vehicle is small. Therefore, the received power value of the radar installed in the lane in which the vehicle is traveling is weaker than that in the lane where the vehicle is not traveling. Therefore, the lane in which the vehicle is traveling can be specified based on the magnitude of the received power value of the radar.

  This embodiment can be used particularly for specifying a lane on a curved road. FIG. 15 is a diagram for explaining an example of a combination of a vehicle running state and radar observation results. FIG. 15A is a diagram showing a case where one vehicle H travels on a lane 2 on a curved road. FIG. 15B shows a change with time of the detection distance of the object Hd (object ID = 4) by the radar 11, and FIG. 15D shows a change with time of the received power of the radar 11. FIG. 15C shows a change with time of the detection distance of the object H (object ID = 5) by the radar 12, and FIG. 15E shows a change with time of the received power of the radar 12. On the right curve as shown in FIG. 15A, a vehicle traveling in the lane 2 is installed in the lane 1 at a point closer to the radar than a point where the covered area 4a of the radar 11 and the covered area 4b of the radar 12 overlap. The radar 11 is detected. Therefore, the object H and the object Hb are not significantly different from each other in the detection start time and the initial value of the detection position. However, as shown in FIG. 15D and FIG. 15E, the received power of the radar 12 installed in the lane 2 where the vehicle H is traveling is larger than the received power of the radar 11. Therefore, the determination unit 56 can correctly identify the lane in which the vehicle is traveling by determining that the vehicle is traveling in the lane in which the receiving radar is installed.

  However, when two vehicles run side by side in adjacent lanes as shown in FIGS. 15 (f) to 15 (j), there is almost no difference in received power between FIGS. 15 (i) and 15 (j). . Even in such a state, if it is determined that one vehicle is traveling in the lane in which the receiving radar is installed, one of the vehicles traveling side by side is lost. Therefore, the determination unit 56 can further use a power difference threshold value to detect a case where two vehicles travel side by side in adjacent lanes as shown in FIGS. 15 (f) to 15 (j). When the difference between the received power values from the target object to be determined is smaller than the power difference threshold, the determination unit 56 detects each of the two vehicles running side by side in the adjacent lane. Judge that there is. Note that the power difference threshold value can be set in advance so that the error rate when the traveling state of the vehicle is determined is smaller than a certain value. Here, the upper limit value of the error rate can be arbitrarily set according to the detection accuracy required for the vehicle detection device 80.

  FIG. 16 is a diagram illustrating an example of a method for determining a power difference threshold value. If the power difference threshold is too small, the vehicle detection device 80 cannot detect one of the vehicles running side by side. Therefore, in order to obtain the lower limit value of the power difference threshold value, the data obtained in the parallel running state as shown in FIG. 16A is processed using various values of the power difference threshold value. For each value used as the power difference threshold, a determination error rate is calculated, and a lower limit value of the power difference threshold is obtained based on the error rate. FIG. 16B is an example of an experimental result showing the relationship between the number of times that the determination was erroneous when the evaluation was performed 28 times and the value of the power difference threshold value used for the determination. FIG. 16C is a graph showing the error rate shown in FIG. 16B as a function of the magnitude of the power difference threshold. In this experimental example, the error rate was 4% when the power difference threshold was 3.0 dBmV, and the error rate was 0% when 6.0 dBmV.

  On the other hand, if the power difference threshold is too small, even when only one vehicle is traveling, it is mistaken that there is a vehicle traveling in parallel. Therefore, in order to obtain the upper limit value of the power difference threshold, as shown in FIG. 16 (d), data obtained in a state where one vehicle is traveling is used by using various values of the power difference threshold. It is processed. FIG. 16E is an example of an experimental result showing the relationship between the number of times that the determination was erroneous when 29 evaluations were performed and the value of the power difference threshold value used for the determination. FIG. 16F is a graph showing the error rate shown in FIG. 16E as a function of the magnitude of the power difference threshold. In this experimental example, the error rate was 0% until the power difference threshold was 6.2 dBmV.

  The operator or the like of the vehicle detection device 80 can obtain the value of the power difference threshold value based on the result of performing a similar experiment at the point where the vehicle detection device 80 is installed. Note that the method shown in FIG. 16 is an example of how to obtain the power difference threshold value, and other calculation methods and experimental methods may be changed as appropriate by an operator or the like.

  FIG. 17 is a flowchart for explaining an example of the operation of the control unit 81. In the example of FIG. 17, an example of determination on an object detected by the radar 11 installed in the lane 1 and an object detected by the radar 12 installed in the lane 2 is shown. Note that the control unit 81 can similarly process objects detected in the other two adjacent lanes. The processes of steps S31 and S32 are the same as steps S11, S12, and S16 described with reference to FIG. The determination unit 56 compares the difference between the weighted average values of received power with a power difference threshold value (step S33). When the difference between the weighted average values of the received power is equal to or greater than the power difference threshold, the determination unit 56 determines that the received power value of the target detected by the radar 11 is higher than the received power of the target detected by the radar 12. It is confirmed whether it is larger (step S34). The determination unit 56 determines that the vehicle is located in the lane 1 when the received power value of the target detected by the radar 11 is larger (step S35). On the other hand, when the received power value of the object detected by the radar 11 is smaller, the determination unit 56 determines that the vehicle is located in the lane 2 (step S36). Further, when the difference between the weighted average values of the received power is less than the power difference threshold, the determination unit 56 determines that the two objects have detected each of the vehicles running side by side (step S37).

<Fourth Embodiment>
It is also possible to combine the first to third embodiments. Hereinafter, an example in which the first embodiment and the third embodiment are combined will be described.

  FIG. 18 is a diagram illustrating an example of the configuration of the vehicle detection device 90 according to the fourth embodiment. The vehicle detection device 90 determines the lane in which the vehicle is located based on the received power value when the detection distance is less than the second threshold, and detects the detection distance when processing an object with the detection distance equal to or greater than the second threshold. The lane in which the vehicle is located is determined using the initial value. The vehicle detection device 90 includes a communication unit 31 (31a to 31c), a memory 32, a detection unit 33 (33a to 33c), a transmission unit 34, and a control unit 91. The control unit 91 includes an acquisition unit 41, an integration unit 42, an output unit 43, a target extraction unit 51, a distance comparison unit 52, a distance observation radar identification unit 53, a reception radar identification unit 55, and a determination unit 56. Operations other than the distance comparison unit 52 and the determination unit 56 are the same as those in the first or third embodiment. However, the target extraction unit 51 also notifies the distance comparison unit 52 of the combination of objects to be determined.

  The distance comparison unit 52 compares the distance between the position of the target object to be determined and the radar with a second threshold value. The second threshold value can be, for example, a length up to half the total length of the radar coverage. In the example of FIG. 1, since the radar coverage ranges from 42 to 122 m, the second threshold can be set to 82 m, for example.

  When the combination of the objects to be determined is notified from the object extracting unit 51, the distance comparing unit 52 compares the detection distance of one of the objects of the combination with the second threshold, and the comparison result is determined by the determining unit. 56 is notified. For example, in the case of the combination of the target object A and the target object Ad, the detection distance of the target object A is 106.5 m. Therefore, the distance comparison unit 52 determines the detection distance of the combination of the target object A and the target object Ad as the second distance. The determination unit 56 is notified that it is larger than the threshold value. On the other hand, in the case of the combination of the target object Cd and the target object C, the detection distance of the target object C is 61.5 m, so the distance comparison unit 52 determines the detection distance of the combination of the target object Cd and the target object C as the second. The determination unit 56 is notified that the value is smaller than the threshold value.

  Further, the determination unit 56 is notified of the reception radar and the power difference threshold value from the reception radar specifying unit 55. The determination unit 56 is notified of the distance observation radar and the distance difference threshold from the distance observation radar specifying unit 53.

  Based on the notification from the distance comparison unit 52, the determination unit 56 determines which of the received power value and the initial value of the detected distance is used to determine the lane in which the vehicle is located. For example, since the detection distance of the target object A or the target object Ad is larger than the second threshold value, the determination unit 56 uses the information for identifying the distance observation radar and the distance difference threshold value to determine the target object A and the target object Ad. The lane where the corresponding vehicle is located is determined. The determination method is the same as in the first embodiment. On the other hand, since the detection distance of the target object Cd and the target object C is smaller than the second threshold value, the determination unit 56 corresponds to the target object Cd and the target object C using the information for identifying the reception radar and the power difference threshold value. The lane where the vehicle to be located is determined. The determination method is the same as in the third embodiment.

  FIG. 19 is a flowchart for explaining an example of the operation of the control unit 91. In the example of FIG. 19, an example of determination of an object detected by the radar 11 installed in the lane 1 and an object detected by the radar 12 installed in the lane 2 is shown. Note that the control unit 91 can similarly process objects detected in the other two adjacent lanes. The processes in steps S41 and S42 are the same as steps S11, S12, and S16 described with reference to FIG. In step S43, the distance comparison unit 52 compares the detection distance of the target object with the second threshold value. When the detection distance of the object is less than the second threshold value, the processes of steps S44 to S48 are performed. The processing in steps S44 to S48 is the same as the processing in steps S33 to S37 described with reference to FIG. On the other hand, when the detection distance of the object is greater than or equal to the second threshold value, the processes of steps S49 to S51 are performed. The process of steps S49 to S51 is the same as the process of steps S13 to S15 described with reference to FIG.

  Thus, by combining a plurality of embodiments, it is possible to detect a vehicle that matches the shape of the road on which the vehicle detection device 90 is installed. For example, on a road having a curve up to the midpoint of the covered area 4, the determination using the received power value may be more effective than the determination using the initial value of the detection distance or the detection start time. Also, when the radar coverage is far from the radar installation position, the received power is small, so the determination using the initial detection distance and detection start time is more effective than the determination based on the received power. There is. Further, the combination can be changed in accordance with the configuration of the apparatus to be mounted, the processing capability, and the like. For example, the time detection radar specifying unit 54 may be added to the vehicle detection device 90 to combine the first to third embodiments. FIG. 20 is a flowchart for explaining an example of an operation when the first to third embodiments are combined. In the flowchart of FIG. 20, when the detection distance is equal to or smaller than the second threshold, the determination is performed based on the received power value. In addition, when the detection distance is greater than the second threshold and less than or equal to the third threshold, the determination is performed using the initial value of the detection distance, and when the detection distance is greater than the third threshold, the detection start time is used. The third threshold value is larger than the second threshold value, and can be arbitrarily changed according to the processing capability of the vehicle detection device. The processing in steps S61 to S68 is the same as that in steps S41 to S48 described with reference to FIG. In step S <b> 69, the distance comparison unit 52 compares the third threshold value with the magnitude of the detection distance, and outputs the comparison result to the determination unit 56. When the detection distance of the target is greater than or equal to the second threshold and less than the third threshold, the processes of steps S70 to S72 are performed. Steps S70 to S72 are the same as steps S13 to S15 described with reference to FIG. On the other hand, if the value is greater than or equal to the third threshold value, the processes of steps S73 to S75 are performed. Steps S73 to S75 are the same as steps S23 to S25 described with reference to FIG. Note that the second embodiment and the third embodiment may be combined.

<Fifth Embodiment>
In the fifth embodiment, an example of the operation of the vehicle detection device 90 will be described. In the following description, processing performed after the data processing at time t0 ends and the vehicle detection device 90 receives the data at time t1 will be described. In addition, the following description is an example, Comprising: The acquisition part 41, the integration part 42, the output part 43, the object extraction part 51, the distance comparison part 52, the distance observation radar specific part 53, the receiving radar specific part 55, or the determination part The operation of 56 can be changed according to the implementation. For example, the integration unit 42 can omit the processes in steps S89 to S92. In the present embodiment, the invalid value of the object ID or the vehicle ID is “−1”, but any other value can be set in advance as an invalid value. Furthermore, in this embodiment, the case where a process is performed from the left lane is described. In this embodiment, both the input table and the integrated data table are stored in the memory 32.

  The processes of the communication unit 31 and the detection unit 33 are the same as those in the first embodiment. However, in the present embodiment, the detection unit 33 notifies the control unit 91 that the detection data table has been generated. FIG. 21 is a diagram illustrating an example of a detection data table and an input table at time t1. FIG. 21A shows the detection unit 33a, FIG. 21B shows the detection unit 33b, and FIG. 21C shows the detection data table at time t1 generated by the detection unit 33c. As shown in FIG. 1 (b), the vehicle A is not detected because it has escaped from both the covered area 4 a of the radar 11 and the covered area 4 b of the radar 12. Accordingly, the radar 11 detects the object B and the object Cd corresponding to the vehicle B and the vehicle C. The radar 12 detects the object Bd, the object C, and the object E corresponding to the vehicles B, C, and E because the vehicle E newly enters the covered area 4b. The radar 13 detects the object D corresponding to the vehicle D. The acquisition unit 41 changes the data included in the input table at time t0 shown in FIG. 5 according to the detection data tables in FIGS. 21A to 21C, and sets the marker to 0. Here, the marker is used to confirm whether the detected object is used when the integrated data table is generated. It is assumed that the marker value of the data used for generating the integrated data table is “1” and the marker value of the unused data is “0”. The acquisition unit 41 generates an input data table shown in FIG. The method for changing the data in the input data table is the same as the method described with reference to FIG.

  When the control unit 91 receives a notification that the detection data table has been generated, the control unit 91 notifies the integration unit 42 of the notification. The integration unit 42 sets the merge flag of the integrated data table (FIG. 9) to 0 as shown in FIG. 22A when the processing at time t0 is completed. The data for one line of the integrated data table may be described as “integrated data”. Further, the data in the first row in FIG. 22A may be described as integrated data of vehicle ID = 1 (vehicle A).

  The merge flag indicates whether there is a possibility that certain integrated data may be further integrated with data such as other objects. When there is a possibility that the integrated data is further integrated, the merge flag is “0”. When there is no possibility that the integrated data is integrated, the merge flag is “1”.

  FIG. 23 is a flowchart for explaining an example of a comparison process between an object in the integrated data table and an object in the input table. FIG. 23A is a flowchart for explaining the presence confirmation of the object, and FIG. 23B is a flowchart for explaining the change of the integrated data table based on the result of FIG. The integration unit 42 sets the variable y to 1 (step S81a). The integration unit 42 sets the variable n to 1 and sets the object presence flag to 0 (step S81b). Here, n is a variable for counting the number of radars, and y is a variable for counting the number of processed data. N is a constant indicating the number of radars, and Y is a constant indicating the number of integrated data recorded in the integrated data table. The integration unit 42 confirms whether the object ID of the nth radar for the yth integrated data is “−1” (step S82). When the object ID is not “−1”, it is determined whether the same object is repeatedly detected based on information from the nth radar in the input table (steps S83 and S84). For example, in the integrated data of vehicle ID = 1 shown in FIG. 22A, the object ID of the object detected by the radar 11 is 1. Therefore, the integration unit 42 confirms the input table in FIG. 21D and confirms whether the object having the object ID = 1 is detected by the radar 11. When the object is detected, the integration unit 42 sets the object presence flag to 1 (step S85). Further, the integration unit 42 records 1 in the marker column of the input table for the confirmed object (step S86). On the other hand, when the object is not detected, the integration unit 42 sets the object ID of the integrated data table to “−1” (step S87). For example, the object with the object ID = 1 of the radar 11 is not recorded in the input table of FIG. Therefore, as shown in FIG. 22B, the object ID in the radar 11 of the integrated data of the vehicle ID = 1 is set to “−1”. On the other hand, when the integration unit 42 confirms whether there is an object with the object ID = 2 of the radar 11, the object with the object ID = 2 of the radar 11 is recorded in the input table. Therefore, the integration unit 42 sets the marker of the object ID = 2 detected by the radar 11 to 1, as shown in FIG. Thereafter, the integration unit 42 increments n by 1, compares it with N, and repeats the processes of steps S82 to S88 until n becomes larger than N (step S88).

  Next, when the object presence flag is 1, the integration unit 42 increments the number of continuations by 1 for the vehicle (steps S89 and S90). On the other hand, if the object presence flag is 0, the integration unit 42 sets the vehicle ID of the yth vehicle to “−1” and sets the number of continuations to 0 (steps S91 and S92). Here, the number of continuations is the number of times the same vehicle is continuously detected. The integration unit 42 increments y by 1 and then compares it with Y, and repeats the processing of steps S81b to S93 until y becomes larger than Y (step S93).

  When the process of FIG. 23A ends, the integration unit 42 checks the vehicle ID of each integrated data. If the vehicle ID is “−1”, there is no object detected corresponding to the vehicle, and therefore data relating to the vehicle is deleted (steps S95 to S97). FIG. 24 is a diagram illustrating an example of the integrated data table after data related to a vehicle in which a corresponding target object is not detected. In FIG. 24, the information of the vehicle A is deleted from FIG.

  25A and 25B are flowcharts illustrating an example of a method for generating an integrated data table. When the comparison process between the object in the integrated data table and the object in the input table is completed, the integrating unit 42 sets the variable y to 1 again (step S101a). Furthermore, the integration unit 42 sets the variable n to 1 (step S101b). If the object ID of the object detected by the nth radar in the yth integrated data is not −1, the integration unit 42 confirms the representative radar number (steps S102 and S103). The representative radar is a radar having a smaller radar number among radars that detect the object determined by the determination unit 56. In the present embodiment, as shown in FIG. 1 and the like, the radar installed in the left lane has a lower number. If the representative radar number is -1, the integrating unit 42 sets the nth radar as the representative radar and increments the value of n (steps S104 and S105). On the other hand, when the representative radar is already set or when the object ID of the object detected by the nth radar in the integrated data of the yth vehicle is −1, the integration unit 42 Is incremented (step S105). The integration unit 42 repeats steps S102 to S105 until the value of n becomes larger than N.

  When the value of n becomes larger than N, the integration unit 42 sets the target object data detected by the representative radar for the yth integrated data to the yth integrated data (steps S106 and S107). An example when the setting to the integrated data is finished is shown in vehicle ID = 3, 4 of FIG. The integration unit 42 confirms the number of object IDs that are not invalid values (−1) in the y-th integrated data, and transitions to step S121 when there is one valid object ID (step S108). On the other hand, when the plurality of object IDs have valid values, the integration unit 42 sets the variable z to 1 (step S109). The variable z is used to count the number of radars.

  The integration unit 42 confirms whether the z-th radar is a representative radar (step S110). When the z-th radar is not the representative radar, it is confirmed whether or not the object ID of the z-th radar is −1 in the y-th integrated data (step S111). When the object ID is not −1, the integration unit 42 acquires data associated with the object ID of the z-th radar for the y-th vehicle from the input table and notifies the determination unit 56 of the data. (Step S112). An example of the determination process will be described later. The determination process can be the same as the determination described in the first to fourth embodiments.

  Based on the output result of the determination unit 56, the integration unit 42 confirms whether the object detected by the representative radar and the object detected by the z-th radar are integrated into the detection result of one vehicle (step) S113). When it is not integrated into one vehicle, the integration unit 42 sets the merge flag of the y-th integrated data to 0 and keeps the y-th integrated data in a state where it can be integrated with other objects (step S114). Further, the integration unit 42 sets the marker of the input table to 0 for the object on the z-th radar used for the determination process (step S115). Further, the integration unit 42 sets the object ID of the yth integrated data on the zth radar to −1 (step S116).

  On the other hand, if the object used for the determination corresponds to one vehicle, the integration unit 42 integrates the data of the object detected by the zth radar with the yth integrated data. Is confirmed (step S117). When integrating with the y-th integrated data, the integrating unit 42 sets the merge flag of the y-th integrated data to 1 (step S118). On the other hand, when not integrated with the y-th integrated data, the integration unit 42 sets the merge flag of the y-th integrated data to 0 (step S119). Further, the integration unit 42 performs integration processing of integrated data. The integration process will be described later. When the integration process ends, the integration unit 42 increments the value of z (step S120). Steps S110 to S120 are repeated until the value of z becomes greater than N. When z becomes larger than N, the integration unit 42 increments the value of y (step S121). Steps S101b to S121 are repeated until the value of y becomes larger than Y.

  Note that the method of setting the merge flag in S117, S118, and S119 is not limited to this method. For example, even if the target detected by the z-th radar is integrated with the y-th integrated data in S117, another radar adjacent to the representative radar (input 1) is replaced with the z-th radar (input 2). ), And in the integrated data being processed, if the object ID corresponding to the other radar is “−1”, the merge flag may be left at 0. This is because, depending on the shape of the road where the radar is installed and the traffic state of the vehicle at the time of detection, the data of the other radar can be further integrated with the data integrated into the input 1 at a later time. This is because there may be sex.

  27A and 27B are flowcharts for explaining an example of the determination process. 27 is performed by the determination unit 56. FIG. 27A shows that the determination process includes determination and determination of a determination method. FIG. 27B shows a determination method determination method, and FIG. 27C shows a determination method.

  The determination unit 56 determines whether two objects are detected by adjacent radars (step S131). When it is not detected by the adjacent radars, the determination unit 56 does not perform determination. On the other hand, when the two objects are detected by the adjacent radars, it is confirmed whether the distance between the two objects is less than the first threshold (step S132). When the distance between the two objects is equal to or greater than the first threshold, the determination unit 56 does not perform determination. When the distance between the two objects is less than the first threshold, it is confirmed whether the detected distance is less than the second threshold (step S133). When the detection distance is less than the second threshold, the determination unit 56 performs determination based on received power. On the other hand, when the detection distance is equal to or greater than the second threshold, the determination unit 56 checks whether the difference between the initial values of the detection distances of the two objects is equal to or greater than the distance difference threshold (step S134). When the difference in the initial value of the detection distance is greater than or equal to the distance difference threshold, the determination unit 56 performs determination based on the initial value of the detection distance. On the other hand, when the difference between the initial values of the detection distances is less than the distance difference threshold, the determination unit 56 determines that the two objects are side-by-side vehicles and ends the determination process. In this case, the return value is “not integrated into one unit”.

  Next, the determination process will be described. The determination unit 56 confirms the determination method (step S141). In the case of determination based on the received power, the determination unit 56 checks whether the difference between the weighted average values of the received power is less than the power difference threshold value (step S142). When the difference between the weighted average values of the received power is less than the power difference threshold, the determination unit 56 determines that the two objects are side-by-side vehicles, and sets the return value to “not integrated into one”. On the other hand, when the difference between the weighted average values of the received power is greater than or equal to the power difference threshold, the determination unit 56 checks whether the weighted power value of the target object 1 is higher than the weighted power value of the target object 2 (step S143). . When the weighted power value of the target object 1 is higher, the determination unit 56 sets the return value to “integrated with the target object 1”. On the other hand, when the weighted power value of the target object 1 is lower, the determination unit 56 sets the return value as “integrated with the target object 2”. When determining with the initial value of the distance, the determination unit 56 checks whether the initial value of the detection distance of the object 1 is smaller than the initial value of the detection distance of the object 2 (step S144). When the initial value of the detection distance of the target object 1 is smaller, the determination unit 56 sets the return value as “integrated with the target object 1”. On the other hand, when the initial value of the detection distance of the object 2 is smaller, the determination unit 56 sets the return value as “integrated with the object 2”. Further, when the determination is not performed, the determination unit 56 sets the return value to “not integrated into one unit”. The determination unit 56 outputs the return value to the integration unit 42.

  FIG. 28 is a flowchart for explaining an example of the integration process. The integration unit 42 confirms the return value acquired from the determination unit 56 (step S151). When the return value is “integrated with the object 1”, the integration unit 42 sets the data of the object 1 in the integrated data (step S152). When the return value is “integrated with the object 2”, the integration unit 42 sets the data of the object 2 in the integrated data (step S153). An example of the integrated data after the processing of FIGS. 25 to 28 is shown in FIG.

  Note that the determination in S134 may be performed between S141 and S144. In that case, the branch of No in S133 is directly connected to “determination based on the initial value of the detection distance”, and the determination in S134 is inserted between S141 and S144. Then, the branch destination of No in S134 is connected to “END return value: not integrated into one”, and the branch destination of Yes is connected to S144.

  FIG. 29 is a flowchart illustrating an example of a new registration method. The integration unit 42 uses a variable u and a constant U. u is used to count the number of processed objects. U is the number of objects whose markers in the input table are 0 when new registration is started. When starting the new registration, the integration unit 42 refers to the input table, acquires the value of U, and sets u to 1 (step S161). The integration unit 42 acquires data and a radar number for the u-th object whose marker is 0 (step S162). The integration unit 42 sets the value of y to 1 (step S163).

  The integration unit 42 confirms whether there is a possibility that the y-th integrated data is integrated with the u-th object (step S164). That is, the integration unit 42 confirms whether the merge flag of the yth integration data is 0. Furthermore, the integration unit 42 confirms that the y-th integrated data is not already integrated with another object detected by the same radar as the radar that detected the u-th object. If the integration is possible, the integration unit 42 notifies the determination unit 56 to that effect and receives the determination result (step S165). The determination process is the same as the method described with reference to FIG. Steps S166 to S168 are the same as steps S117 to S119 described with reference to FIG. 25B. The integration process is as described with reference to FIG. In step S169, the integration unit 42 records the u-th object ID in the integrated data table. Thereafter, the integration unit 42 increments y and compares it with Y (step S170). Also, when it is determined No in steps S164 and S165, the integration unit 42 increments y and compares it with Y. Steps S164 to S170 are repeated until y becomes larger than Y.

  When the u-th object is not integrated with any integrated data, the integration unit 42 newly registers the data of the u-th object in the integrated data table and sets the merge flag to 0 (steps S171 to S173). ). Further, the integration unit 42 sets the marker of the u-th object in the input table to 1, and increments the value of u by 1 (steps S174 and S175). Even when any of the integrated data and the u-th object are integrated, the processes of steps S174 and S175 are performed. Steps S162 to S175 are repeated until u becomes larger than U. FIG. 26B shows an example of the integrated data table in which the vehicle E is registered in the new registration process, and FIG. 26C shows an example of the input table. FIG. 30 is a flowchart illustrating an example of determination between integrated data. The integration unit 42 confirms whether there is a plurality of integrated data corresponding to the same vehicle. In the process of FIG. 30, the integration unit 42 uses the variable i. The yth integrated data is compared with the y + ith integrated data.

  First, the integration unit 42 sets the variable y to 1 (step S180). The integration unit 42 confirms whether the y-th integrated data can be integrated with other data. That is, the integration unit 42 checks whether the vehicle ID is −1 and the merge flag is 1 for the y-th integrated data (step S181). When the y-th integrated data can be integrated with other data, the integrating unit 42 identifies the radar that observed the object associated with the y-th integrated data (step S182). The integration unit 42 sets the variable i to 1 (step S183a). Next, the integration unit 42 performs the same processing as steps S181 and S182 on the y + i-th integration data (steps S183b and S184). Next, the integration unit 42 confirms whether there is an object observed by the same radar in the y-th integrated data and the y + i-th integrated data (step S185). If there is no object observed by the same radar, the integration unit 42 notifies the determination unit 56 to that effect, and the determination unit 56 performs a determination process. Steps S186 to S189 are the same as steps S165 to S168 described with reference to FIG. The integration unit 42 sets the variable n to 1 (step S190a). Further, when the object ID by the nth radar in the yth integrated data is −1, the integration unit 42 substitutes the object ID by the nth radar in the y + ith integrated data (step S190b). . The integration unit 42 increments n by 1 and repeats steps S190b and S191 until n becomes larger than N.

  Further, the integration unit 42 performs integration processing, sets the vehicle ID of the y + i-th data to −1, and increments i by 1 (steps S192 and S193). Also, in the case of Yes in steps S183b and S185 and in the case of No in S186, the integration unit 42 increments i by 1. Steps S183b to S193 are repeated until i becomes larger than Y-2. When i becomes larger than Y-2, the integration unit 42 increments y by 1 and compares it with Y-1 (step S194). Also, if it is determined Yes in step S181, the process of step S194 is performed. Steps S181 to S194 are repeated until y is greater than Y-1.

<Others>
The present invention is not limited to the above-described embodiment, and can be variously modified. Some examples are described below.

  The objects detected corresponding to one vehicle are considered to be similar in position and speed. Therefore, in the sixth embodiment, when the distance between two objects is less than the first threshold and the difference in speed is equal to or less than a predetermined speed threshold, the two objects are one. It is determined that there is a possibility that the vehicle is, and the detection process is performed. As shown in FIG. 4, the detection data table includes the moving speed of the object. Therefore, the target extraction unit 51 designates two objects whose positions and speeds are similar to each other as processing targets when two objects that are similar to each other are detected by a radar installed in an adjacent lane. You can also

The processing from the left lane is an example of the processing method, and the processing order can be changed according to the implementation, such as processing from the right lane.
When the vehicle detection device 20 is realized by a computer, the Central Processing Unit (CPU) operates as the communication unit 31, the detection unit 33, the control units 40, 71, 81, 91, and the transmission unit 34 by reading a program. . The program is stored in the memory 32.

The following additional notes are further disclosed for each of the embodiments described above.
(Appendix 1)
A vehicle detection device that is installed in each of a plurality of lanes and detects a vehicle using information from a radar that includes a lane adjacent to the installation lane in a detection range,
From each radar, a detection distance representing a distance between the detected object and the radar that detected the object, a detection time of the object, and an acquisition unit that acquires received power from the object;
The distance between the first object detected by the first radar and the second object detected by the second radar installed in the lane adjacent to the lane where the first radar is installed is If the first threshold is less than or equal to the first threshold, the latest detection distance of the first object is compared with a second threshold, and the latest detection distance is compared with a third threshold that is greater than the second threshold. A distance comparison unit,
A distance observation radar specifying unit that specifies a distance observation radar that has observed the shortest distance among the detection distance of the first object and the detection distance of the second object;
Of the detection time of the first object and the detection time of the second object, a time observation radar specifying unit for specifying a time observation radar that observed the earliest detection time;
A receiving radar specifying unit for specifying a receiving radar that has received the largest received power among the received power from the first and second objects;
When the latest detection distance is greater than or equal to the third threshold, it is determined that the vehicle is traveling in the lane where the time observation radar is installed, and the latest detection distance is less than the third threshold and the If it is equal to or greater than the second threshold, it is determined that the vehicle is traveling in the lane where the distance observation radar is installed, and if the latest detection distance is less than the second threshold, the reception radar is installed A vehicle detection device comprising: a determination unit that determines that the vehicle is traveling in the lane in which the vehicle is running.
(Appendix 2)
A vehicle detection device that is installed in each of a plurality of lanes and detects a vehicle using information from a radar that includes a lane adjacent to the installation lane in a detection range,
From each radar, a detection distance that represents the distance between the detected object and the radar that detected the object, and an acquisition unit that acquires the detection time of the object;
Detection distance of the first object detected by the first radar and detection of the second object detected by the second radar installed in the lane adjacent to the lane where the first radar is installed Specify the distance observation radar that observed the shortest distance among the distances, or the time observation radar that observed the earliest detection time of the detection time of the first object and the detection time of the second object An observation radar identification unit to
When the distance between the first object and the second object is equal to or less than a first threshold, the vehicle travels on the time observation radar or a lane in which the distance observation radar is installed. A vehicle detection device comprising: a determination unit that determines that the vehicle is present.
(Appendix 3)
A vehicle detection device that is installed in each of a plurality of lanes and detects a vehicle using information from a radar that includes a lane adjacent to the installation lane in a detection range,
From each radar, a detection distance that represents the distance between the detected object and the radar that detected the object, and an acquisition unit that acquires received power from the object;
The distance between the first object detected by the first radar and the second object detected by the second radar installed in the lane adjacent to the lane where the first radar is installed is A reception radar specifying unit that specifies a reception radar that has received the largest received power among the received power from the first and second objects when the first threshold value or less;
A vehicle detection device comprising: a determination unit that determines that a vehicle is traveling in a lane in which the reception radar is installed.
(Appendix 4)
A distance comparison unit that compares the latest detection distance of the first object with a second threshold;
A distance observation radar specifying unit that specifies a distance observation radar that has observed the shortest distance among the detection distance of the first object and the detection distance of the second object;
The determination unit determines that the vehicle is traveling in a lane in which the distance observation radar is installed when the latest detection distance is equal to or greater than the second threshold, and the latest detection distance is the second detection distance. The vehicle detection device according to appendix 3, wherein the vehicle detection device determines that the vehicle is traveling in a lane where the reception radar is installed.
(Appendix 5)
The determination unit detects a detection distance when the first object is first detected by the first radar and a detection when the second object is first detected by the second radar. Compare the detected distance difference value representing the difference in distance with a predetermined distance difference threshold,
When the detected distance difference value is smaller than the distance difference threshold, it is determined that the vehicle is traveling in both the lane where the first radar is installed and the lane where the second radar is installed. The vehicle detection device according to appendix 1, 2, or 4, characterized in that:
(Appendix 6)
The determination unit calculates a difference between a detection time when the first object is first detected by the first radar and a detection time when the second object is first detected by the second radar. Comparing the detected detection time difference value to a predetermined time threshold,
If the detected time difference value is smaller than the time threshold value, it is determined that the vehicle is traveling in both the lane in which the first radar is installed and the lane in which the second radar is installed. The vehicle detection device according to appendix 1, 2, or 4, wherein
(Appendix 7)
The determination unit obtains a received power difference value representing a difference between received power received by the first radar from the first object and received power received by the second radar from the second object. Compared to a predetermined power difference threshold,
When the received power difference value is smaller than the power difference threshold value, it is determined that the vehicle is traveling in both the lane in which the first radar is installed and the lane in which the second radar is installed. The vehicle detection device according to appendix 1, 3 or 4, characterized in that:
(Appendix 8)
A first radar installed in the first lane;
A second radar installed in a second lane adjacent to the first lane;
From each radar, an acquisition means for acquiring a detection distance representing a distance between a detected object and a radar that detects the object, a detection time of the object, and received power from the object;
When the distance between the first object detected by the first radar and the second object detected by the second radar is less than or equal to a first threshold, the first object A distance comparison means for comparing the latest detected distance with a second threshold, and comparing the latest detected distance with a third threshold greater than the second threshold;
A distance observation radar specifying means for specifying a distance observation radar that has observed the shortest distance among the detection distance of the first object and the detection distance of the second object;
Time observation radar specifying means for specifying a time observation radar that observed the earliest detection time among the detection time of the first object and the detection time of the second object;
A receiving radar specifying means for specifying a receiving radar that has received the largest received power among the received power from the first and second objects;
When the latest detection distance is greater than or equal to the third threshold, it is determined that the vehicle is traveling in the lane where the time observation radar is installed, and the latest detection distance is less than the third threshold and the If it is equal to or greater than the second threshold, it is determined that the vehicle is traveling in the lane where the distance observation radar is installed, and if the latest detection distance is less than the second threshold, the reception radar is installed A vehicle detection system comprising: determination means for determining that the vehicle is traveling in the lane in which the vehicle is running.

4 Covered area 11, 12, 13 Radar 20, 70, 80, 90 Vehicle detection device 31 Communication unit 32 Memory 33 Detection unit 34 Transmission unit 40, 71, 81, 91 Control unit 41, 44 Acquisition unit 42 Integration unit 43 Output unit 51 Target Extraction Unit 52 Distance Comparison Unit 53 Distance Observation Radar Identification Unit 54 Time Observation Radar Identification Unit 55 Reception Radar Identification Unit 56 Determination Unit 60 Host Device 61 Optical Beacon

Claims (5)

A vehicle detection device that is installed in each of a plurality of lanes and detects a vehicle using information from a radar that includes a lane adjacent to the installation lane in a detection range,
From each radar, a detection distance representing a distance between the detected object and the radar that detected the object, a detection time of the object, and an acquisition unit that acquires received power from the object;
The distance between the first object detected by the first radar and the second object detected by the second radar installed in the lane adjacent to the lane where the first radar is installed is If the first threshold is less than or equal to the first threshold, the latest detection distance of the first object is compared with a second threshold, and the latest detection distance is compared with a third threshold that is greater than the second threshold. A distance comparison unit,
A distance observation radar specifying unit that specifies a distance observation radar that has observed the shortest distance among the detection distance of the first object and the detection distance of the second object;
Of the detection time of the first object and the detection time of the second object, a time observation radar specifying unit for specifying a time observation radar that observed the earliest detection time;
A receiving radar specifying unit for specifying a receiving radar that has received the largest received power among the received power from the first and second objects;
When the latest detection distance is greater than or equal to the third threshold, it is determined that the vehicle is traveling in the lane where the time observation radar is installed, and the latest detection distance is less than the third threshold and the If it is equal to or greater than the second threshold, it is determined that the vehicle is traveling in the lane where the distance observation radar is installed, and if the latest detection distance is less than the second threshold, the reception radar is installed A vehicle detection device comprising: a determination unit that determines that the vehicle is traveling in the lane in which the vehicle is running. A vehicle detection device that is installed in each of a plurality of lanes and detects a vehicle using information from a radar that includes a lane adjacent to the installation lane in a detection range,
From each radar, a detection distance that represents the distance between the detected object and the radar that detected the object, and an acquisition unit that acquires the detection time of the object;
Detection distance of the first object detected by the first radar and detection of the second object detected by the second radar installed in the lane adjacent to the lane where the first radar is installed Specify the distance observation radar that observed the shortest distance among the distances, or the time observation radar that observed the earliest detection time of the detection time of the first object and the detection time of the second object An observation radar identification unit to
When the distance between the first object and the second object is equal to or less than a first threshold, the vehicle travels on the time observation radar or a lane in which the distance observation radar is installed. A vehicle detection device comprising: a determination unit that determines that the vehicle is present. A vehicle detection device that is installed in each of a plurality of lanes and detects a vehicle using information from a radar that includes a lane adjacent to the installation lane in a detection range,
Before acquiring a detection distance representing the distance between the detected object and the radar that detected the object, and the received power from the object, and performing a new acquisition of the received power, from each radar An acquisition unit that calculates a weighted average value of the received power acquired at the time of acquisition and the received power acquired by the new acquisition ;
The distance between the first object detected by the first radar and the second object detected by the second radar installed in the lane adjacent to the lane where the first radar is installed is A receiving radar specifying unit that specifies a receiving radar that has received the weighted average value of the received power that is the largest of the weighted average values of the received power from the first and second objects in the case of the first threshold value or less; ,
A vehicle detection device comprising: a determination unit that determines that a vehicle is traveling in a lane in which the reception radar is installed. The determination unit detects a detection distance when the first object is first detected by the first radar and a detection when the second object is first detected by the second radar. Compare the detected distance difference value representing the difference in distance with a predetermined distance difference threshold,
When the detected distance difference value is smaller than the distance difference threshold, it is determined that the vehicle is traveling in both the lane where the first radar is installed and the lane where the second radar is installed. The vehicle detection device according to claim 1, wherein the vehicle detection device is a vehicle detection device. A first radar installed in the first lane;
A second radar installed in a second lane adjacent to the first lane;
From each radar, an acquisition means for acquiring a detection distance representing a distance between a detected object and a radar that detects the object, a detection time of the object, and received power from the object;
When the distance between the first object detected by the first radar and the second object detected by the second radar is less than or equal to a first threshold, the first object A distance comparison means for comparing the latest detected distance with a second threshold, and comparing the latest detected distance with a third threshold greater than the second threshold;
A distance observation radar specifying means for specifying a distance observation radar that has observed the shortest distance among the detection distance of the first object and the detection distance of the second object;
Time observation radar specifying means for specifying a time observation radar that observed the earliest detection time among the detection time of the first object and the detection time of the second object;
A receiving radar specifying means for specifying a receiving radar that has received the largest received power among the received power from the first and second objects;
When the latest detection distance is greater than or equal to the third threshold, it is determined that the vehicle is traveling in the lane where the time observation radar is installed, and the latest detection distance is less than the third threshold and the If it is equal to or greater than the second threshold, it is determined that the vehicle is traveling in the lane where the distance observation radar is installed, and if the latest detection distance is less than the second threshold, the reception radar is installed A vehicle detection system comprising: determination means for determining that the vehicle is traveling in the lane in which the vehicle is running.