Setting support device, information processing method, computer program, and radio wave sensor

Abstract

This setting support device supports setting of a radio wave sensor that detects an object in a detection area by irradiating the detection area with radio waves, the detection area at least partially including a road. The setting support device comprises an acquisition unit that acquires detection results of a plurality of objects by the radio wave sensor, and an identification unit that identifies a low reliability result, which is a detection result having reliability lower than a standard, from the detection results of the plurality of objects acquired by the acquisition unit.

Description

Setting Support Device, Information Processing Method, Computer Program, and Radio Wave Sensor

[0001] The present disclosure relates to a setting support device, an information processing method, a computer program, and a radio wave sensor. This application claims priority based on Japanese Application No. 2024-208248 filed on November 29, 2024, and incorporates all the contents described in the said Japanese application.

[0002] The radio wave sensor is installed at a position where it can detect objects such as vehicles and pedestrians on a road (including intersections) for the purpose of traffic monitoring. Such a radio wave sensor for infrastructure (road facilities) is used, for example, to measure the traffic volume of vehicles traveling on the road and to detect pedestrians on crosswalks. In order to use the radio wave sensor for traffic monitoring, it is necessary to set an area to be detected (hereinafter referred to as the "detection area") such as a lane, a traffic lane, a crosswalk, a sidewalk, etc. in the coordinate system of the radio wave sensor.

[0003] Patent Document 1 discloses a setting support device for assisting a user in setting a detection area of a radio wave sensor.

[0004] International Publication No. 2024 / 122371

[0005] A setting support device according to one aspect of the present disclosure is a setting support device that supports the setting of a radio wave sensor that detects an object in a detection area including at least a part of a road by irradiating radio waves to the detection area, and includes an acquisition unit that acquires detection results of a plurality of objects by the radio wave sensor, and a specifying unit that specifies a low reliability result, which is a detection result with a reliability lower than a reference, from the detection results of the plurality of objects acquired by the acquisition unit.

[0006] Figure 1 is a diagram showing an example of the use of the radio wave sensor according to the first embodiment. Figure 2 is a perspective view showing an example of the external configuration of the radio wave sensor according to the first embodiment. Figure 3 is a diagram illustrating an example of setting the detection area in the sensor coordinate system. Figure 4 is a block diagram showing an example of the hardware configuration of the radio wave sensor according to the first embodiment. Figure 5 is a functional block diagram showing an example of the functions of the radio wave sensor according to the first embodiment. Figure 6 is a flowchart showing an example of the first generation process of reliability data. Figure 7 is a flowchart showing an example of the second generation process of reliability data. Figure 8 is a flowchart showing an example of the third generation process of reliability data. Figure 9 is a block diagram showing an example of the hardware configuration of the setting support device according to the first embodiment. Figure 10 is a functional block diagram showing an example of the functions of the setting support device according to the first embodiment. Figure 11 is a diagram showing a first example of an inappropriate detection area setting. Figure 12 is a diagram showing a second example of an inappropriate detection area setting. Figure 13 is a diagram showing a third example of an inappropriate detection area setting. Figure 14 is a diagram showing an example of the screen display when the detection area setting is not appropriate. Figure 15 is a flowchart showing an example of the setting diagnosis process of the radio wave sensor by the setting support device according to the first embodiment. Figure 16 is a functional block diagram showing an example of the functions of the setting support device according to the second embodiment. Figure 17 is a diagram showing a first example of a radio wave sensor abnormality. Figure 18 is a diagram showing a second example of a radio wave sensor abnormality. Figure 19 is a diagram showing an example of the screen display when the radio wave sensor is abnormal. Figure 20 is a flowchart showing an example of the abnormality diagnosis process of the radio wave sensor by the setting support device according to the second embodiment. Figure 21 is a functional block diagram showing an example of the functions of the setting support device according to the fourth embodiment. Figure 22 is a diagram showing an example of a location where low reliability results appear frequently. Figure 23 is a diagram showing an example of the clustering results of low reliability results. Figure 24 is a flowchart showing an example of the setting support process of the radio wave sensor by the setting support device according to the fourth embodiment. Figure 25 is a functional block diagram showing an example of the functions of the radio wave sensor according to the fifth embodiment. Figure 26 is a flowchart showing an example of the abnormality diagnosis process of the radio wave sensor according to the fifth embodiment.

[0007] For example, radio wave sensors for pedestrian crossings detect pedestrians, and radio wave sensors for roadways detect vehicles; infrastructure radio wave sensors (infrastructure radio wave sensors) have defined objects they are supposed to detect. If a radio wave sensor mistakenly detects an object that is not its intended target (for example, a radio wave sensor for a pedestrian crossing mistakenly detects a vehicle or a tree on the road as a pedestrian), the detection accuracy of the radio wave sensor decreases. Therefore, it is desirable to reduce false detections by radio wave sensors.

[0008] According to this disclosure, information useful for reducing false detections by radio wave sensors can be obtained.

[0009] The embodiments of this disclosure are outlined below.

[0010] (1) The setting support device according to this embodiment is a setting support device that supports the setting of a radio wave sensor that detects objects in a detection area by irradiating the detection area, which includes at least a part of a road, with radio waves, and comprises an acquisition unit that acquires detection results of a plurality of objects by the radio wave sensor, and an identification unit that identifies low reliability results, which are detection results with lower reliability than a standard, from the detection results of the plurality of objects acquired by the acquisition unit. Low reliability results are highly likely to be the result of false detection by the radio wave sensor. Such low reliability results can be used to reduce false detection by the radio wave sensor. Therefore, with the above configuration, low reliability results, which are useful information for reducing false detection by the radio wave sensor, can be acquired.

[0011] (2) The system may further include a generation unit that generates low reliability information regarding the occurrence of low reliability results based on the low reliability results identified by the identification unit in (1) above. This makes it possible to obtain low reliability information, which is useful information for determining the occurrence of false detections.

[0012] (3) In (2) above, the low reliability information may be the number of occurrences of the low reliability detection result per unit time. This makes it possible to obtain low reliability information regarding the frequency of false detections.

[0013] (4) In (2) above, the low reliability information may be the ratio of the number of detection results with low reliability to the total number of detection results for the object. This makes it possible to obtain low reliability information regarding the rate of false detections.

[0014] (5) In any one of (2) to (4) above, the setting support device may further include a first determination unit that determines whether the settings of the radio wave sensor are appropriate based on the low reliability information generated by the generation unit, and a first output unit that outputs the result of the determination by the first determination unit. This makes it possible to notify the user (administrator) that the settings of the radio wave sensor are not appropriate.

[0015] (6) In (5) above, the setting support device further includes a first estimation unit that estimates the cause of the inappropriate setting of the radio wave sensor when the first determination unit determines that the setting of the radio wave sensor is inappropriate, and the first output unit may output the cause estimated by the first estimation unit. This makes it possible to notify the user of the cause of the inappropriate setting of the radio wave sensor.

[0016] (7) In (6) above, the first estimation unit may estimate the cause based on the location where the unreliable detection result occurred. The location where the false detection occurs changes depending on the reason why the radio wave sensor settings are not appropriate. This makes it possible to accurately estimate the reason why the radio wave sensor settings are not appropriate.

[0017] (8) In any one of (2) to (7) above, the setting support device may further include a second determination unit that determines whether the radio wave sensor is abnormal based on the low reliability information generated by the generation unit, and a second output unit that outputs the result of the determination by the second determination unit. This makes it possible to notify the user that the radio wave sensor is abnormal.

[0018] (9) In (8) above, the setting support device further includes a second estimation unit that estimates the cause of the abnormality of the radio wave sensor when the second determination unit determines that the radio wave sensor is abnormal, and the second output unit may output the cause of the abnormality estimated by the second estimation unit. This makes it possible to notify the user of the cause of the abnormality of the radio wave sensor.

[0019] (10) In (9) above, the second estimation unit may estimate the cause of the abnormality based on the location where the unreliable detection result occurred. Depending on the cause of the abnormality of the radio wave sensor, the location where the false detection occurs changes. This makes it possible to accurately estimate the cause of the abnormality of the radio wave sensor.

[0020] (11) In any one of (8) to (10) above, the second determination unit may determine whether the radio wave sensor is abnormal based on the difference between the first low reliability information generated by the generation unit based on the object detection result of the radio wave sensor at the first time point and the second low reliability information generated by the generation unit based on the object detection result of the radio wave sensor at the second time point after the first time point. This makes it possible to accurately determine the occurrence of an abnormality due to deterioration of the radio wave sensor over time.

[0021] (12) In any one of (1) to (11) above, the setting support device may further include a first determination unit that determines candidate areas which are candidates for exclusion areas to be excluded from the detection target of the radio wave sensor based on the detection position of the object in the low reliability result identified by the identification unit, and a third output unit that outputs the candidate areas determined by the first determination unit. This makes it possible to support setting exclusion areas in locations where false detections are likely to occur.

[0022] (13) In the above (12), the setting support device may further include a second determination unit that determines the specified candidate area as the exclusion area when the candidate area output by the third output unit is specified by the user. This makes it possible to determine the exclusion area at an appropriate location specified by the user.

[0023] (14) The information processing method according to this embodiment is an information processing method for processing information about a radio wave sensor that detects objects in a detection area by irradiating the detection area, which includes at least a portion of a road, with radio waves, and includes the steps of: acquiring detection results of a plurality of objects by the radio wave sensor; and identifying low reliability results, which are detection results with lower reliability than a standard, from the acquired detection results of the plurality of objects. With the above configuration, it is possible to acquire low reliability results, which are information useful for reducing false detections by the radio wave sensor.

[0024] (15) The computer program according to this embodiment is a computer program for processing information about a radio wave sensor that detects objects in a detection area by irradiating the detection area, which includes at least a portion of a road, with radio waves, and causes the computer to perform the steps of: acquiring detection results of a plurality of objects by the radio wave sensor; and identifying low-reliability results, which are detection results with lower reliability than a standard, from the acquired detection results of the plurality of objects. With the above configuration, it is possible to acquire low-reliability results, which are useful information for reducing false detections by the radio wave sensor.

[0025] (16) The radio wave sensor according to this embodiment is a radio wave sensor that detects objects in a detection area by irradiating the detection area, which includes at least a part of a road, with radio waves, and comprises a storage unit that stores the detection results of a plurality of objects by the radio wave sensor, and a specification unit that identifies low reliability results, which are detection results with lower reliability than a standard, from the detection results of the plurality of objects stored in the storage unit. With the above configuration, it is possible to obtain low reliability results, which are useful information for reducing false detections by the radio wave sensor.

[0026] This disclosure can be implemented not only as a setting support device having the characteristic configuration described above, an information processing method using characteristic processing as steps, a computer program for causing the setting support device to execute characteristic processing, and a radio wave sensor having the characteristic configuration, but also as a setting method for a radio wave sensor using characteristic processing as steps, or as a computer program for causing a radio wave sensor to execute characteristic processing. Furthermore, this disclosure can be implemented as a semiconductor integrated circuit for part or all of the setting support device, as a system including part of the setting support device for the radio wave sensor, or as a semiconductor integrated circuit for part or all of the control device for the radio wave sensor.

[0027] The embodiments of this disclosure will be described in detail below with reference to the drawings. At least some of the embodiments described below may be combined in any way.

[0028] [1. First Embodiment] [1-1. Radio Wave Sensor] Figure 1 shows an example of the use of a radio wave sensor (infrastructure radio wave sensor) according to the first embodiment. The radio wave sensor 10 according to the first embodiment is a radio wave radar for traffic monitoring and detects pedestrians at a crosswalk 20. The radio wave sensor 10 is, for example, a millimeter-wave radar.

[0029] The pedestrian crossing 20 is provided on the roadway 60 near the intersection 61. Sidewalks 63a and 63b are provided adjacent to the roadway 60. The roadway 60 includes an entry lane 62a for vehicles entering the intersection 61 and an exit lane 62b for vehicles exiting the intersection 61. Sidewalk 63a is adjacent to the entry lane 62a. Sidewalk 63b is adjacent to the exit lane 62b.

[0030] The radio wave sensor 10 is attached to a structure 50 provided on the sidewalk 63b. The structure 50 is several meters high, and the radio wave sensor 10 is installed several meters above the ground. The structure 50 includes, for example, a pole 51 and an arm 52 provided near the upper end of the pole 51, and the radio wave sensor 10 is attached to the arm 52.

[0031] Figure 2 is a perspective view showing an example of the external configuration of a radio wave sensor 10 according to the first embodiment. As shown in Figure 2, the radio wave sensor 10 includes a housing 18 having a transmitting / receiving surface 18a on one side for transmitting and receiving radio waves. The housing 18 houses a transmitting / receiving unit 14 and a detection circuit 17. The transmitting / receiving unit 14 includes a transmitting antenna 15a and a plurality (for example, four) of receiving antennas 16a. The radio wave sensor 10 transmits a modulated wave, which is a radio wave, from the transmitting antenna 15a through the transmitting / receiving surface 18a. The modulated wave hits an object and is reflected, and the receiving antennas 16a receive the reflected wave. The transmitting / receiving unit 14 and the detection circuit 17 perform signal processing on the transmitted wave signal and the received wave signal to detect the distance to the object, the velocity of the object, and the azimuth angle in which the object is located.

[0032] The radio wave sensor 10 detects objects on the crosswalk 20 (e.g., pedestrians, cyclists) by irradiating radio waves (millimeter waves) onto the crosswalk 20 and receiving the reflected waves. Specifically, the radio wave sensor 10 is positioned such that a straight line (hereinafter also referred to as the "projection center axis"), which is a straight line projected vertically onto the ground surface from the center of the transmitting and receiving surface 18a that transmits and receives radio waves, passes through the crosswalk 20. The radio wave sensor 10 can detect the distance from the radio wave sensor 10 to the object, the velocity of the object (velocity in the straight line connecting the radio wave sensor 10 and the object; hereinafter also referred to as the "line of sight velocity"), and the horizontal angle (azimuth angle) of the position where the object is located relative to the projection center axis.

[0033] [1-2. Detection Area] The radio wave sensor 10 has a detection area 30 set, which is the area on the road for detecting objects. In Figure 1, the detection area 30 is shown as if it exists in real space, but the detection area 30 is information set inside the radio wave sensor 10 and is a virtual area.

[0034] For example, a radio wave sensor 10 used to measure the volume of pedestrians and cyclists (including riders; hereinafter, pedestrians and cyclists are simply referred to as "pedestrians") passing through a crosswalk 20, or to control a traffic signal installed at the crosswalk 20, is required to detect not only pedestrians on the crosswalk 20 but also pedestrians waiting to cross on the sidewalk adjacent to the crosswalk 20. For this reason, for example, the detection area 30 includes not only the area of ​​the crosswalk 20 but also the area where pedestrians wait to cross. That is, the detection area 30 may be an area extended from the crosswalk 20 to both sides in the longitudinal direction of the crosswalk 20 (the direction in which pedestrians pass through the crosswalk).

[0035] The radio wave sensor 10 has a coordinate space set up to determine the position of an object. Hereinafter, the coordinate space set up for the radio wave sensor 10 will also be referred to as the "sensor coordinate space." The sensor coordinate space is a coordinate space unique to the radio wave sensor 10.

[0036] In the radio wave sensor 10, a detection area 30 is set in the sensor coordinate space in order to detect objects on the pedestrian crossing 20.

[0037] Figure 3 is a diagram illustrating an example of setting the detection area in the sensor coordinate system. In Figure 3, the point indicated by reference numeral 31O is a point on the ground surface obtained by projecting the installation position of the radio wave sensor 10 vertically downward. The line indicated by reference numeral 31Y is the projection center axis, and the line indicated by reference numeral 31X is a line on the ground surface that intersects the projection center axis 31Y at point 31O.

[0038] For example, the sensor coordinate space is a virtual coordinate space set in the radio wave sensor 10, and is a two-dimensional coordinate space corresponding to the Earth's surface. The sensor coordinate space is defined by the X and Y axes. The origin O of the sensor coordinate space corresponds to the real-world point 31O. The Y axis of the sensor coordinate space corresponds to the real-world projection center axis 31Y. The X axis of the sensor coordinate space corresponds to the real-world line 31X. The sensor coordinate space can be represented not only as an XY Cartesian coordinate system, but also as a polar coordinate system with the origin O as the pole and the Y axis as the initial line.

[0039] In the radio wave sensor 10, a virtual detection area 300 corresponding to the detection area 30 in real space is set in the sensor coordinate space. The detection area 30 is, for example, the area including the pedestrian crossing 20. In the following, the detection area as internal information set in the radio wave sensor 10 is indicated by reference numeral 300, and the detection area in real space that reflects the detection area 300 is also indicated by reference numeral 300.

[0040] For example, the detection area 30 is divided into a zebra area 30_1, which is the area of ​​the pedestrian crossing 20, and waiting areas 30_2A and 30_2B, which are adjacent to the zebra area 30_1.

[0041] Zebra area 30_1 is rectangular. Waiting area 30_2A is adjacent to the first side of zebra area 30_1 that is close to point 31O. Waiting area 30_2B is adjacent to the second side of zebra area 30_1 that is away from point 31O.

[0042] The detection area 300 in the sensor coordinate space is divided into zebra area 300_1 and standby areas 300_2A and 300_2B. Zebra area 300_1 corresponds to zebra area 30_1 in real space. Standby area 300_2A corresponds to standby area 30_2A in real space, and standby area 300_2B corresponds to standby area 30_2B in real space.

[0043] [1-3. Hardware Configuration of Radio Wave Sensor] Figure 4 is a block diagram showing an example of the hardware configuration of a radio wave sensor according to the first embodiment. The radio wave sensor 10 includes a processor 101, a non-volatile memory 102, a volatile memory 103, a transmitting / receiving unit 14, and a communication interface (communication I / F) 107.

[0044] The volatile memory 103 is a semiconductor memory such as SRAM (Static Random Access Memory) or DRAM (Dynamic Random Access Memory). The non-volatile memory 102 is, for example, a flash memory, a hard disk, a ROM (Read Only Memory), etc. The non-volatile memory 102 stores a control program 110 which is a computer program and data used for the execution of the control program 110. Each function of the radio wave sensor 10 is exerted by the control program 110 being executed by the processor 101. The control program 110 can be stored in a recording medium such as a flash memory, a ROM, or a CD-ROM. The processor 101 can detect the position (distance and azimuth angle) and speed of an object according to the control program 110.

[0045] The processor 101 is, for example, a CPU (Central Processing Unit). However, the processor 101 is not limited to the CPU, and may be, for example, a GPU (Graphics Processing Unit), or an ASIC (Application Specific Integrated Circuit), or an FPGA (Field Programmable Gate Array), a CPLD (Complex Programmable Logic Device), or other programmable logic devices. In this case, the ASIC or the programmable logic device is configured to be able to execute the same processing as the control program 110.

[0046] The transmission and reception unit 14 includes a transmission circuit 15 and a reception circuit 16.

[0047] The transmission circuit 15 includes a transmission antenna 15a. Note that the number of the transmission antennas 15a is not limited to one, and may be plural. The transmission circuit 15 generates a modulated wave and transmits the generated modulated wave from the transmission antenna 15a. The transmitted modulated wave hits an object (for example, a pedestrian, a bicycle, a vehicle) and is reflected.

[0048] The receiving circuit 16 includes a receiving antenna 16a. A plurality (four in the figure) of receiving antennas 16a are provided to detect the azimuth angle of an object. The receiving circuit 16 performs signal processing on the received reflected wave. The reflected wave data generated by the signal processing is provided to the processor 101. The processor 101 analyzes the reflected wave data and detects the position (distance and azimuth angle) and velocity of the object.

[0049] The communication I / F 107 can communicate with an external device. The communication I / F 107 is connected to a setting support device 200 (see FIG. 9) via a cable and can transmit the data of the detection result to the setting support device 200. The communication I / F 107 is a wireless communication interface and may be able to communicate with the setting support device 200 wirelessly.

[0050] Setting information 111 is stored in the non-volatile memory 102. The setting information 111 includes the position information of the detection area 300 in the sensor coordinate space and the position information of a mask area described later.

[0051] [1-4. Functions of the radio wave sensor] FIG. 5 is a functional block diagram showing an example of the functions of the radio wave sensor according to the first embodiment. When the processor 101 executes the control program 110, the radio wave sensor 10 functions as a specifying unit 121, a detecting unit 122, a tracking unit 123, a generating unit 124, and an output unit 125.

[0052] The transmission circuit 15 included in the transceiver 14 transmits a transmission signal, which is a modulated wave, from the transmission antenna 15a. The transmission signal from the transmission antenna 15a hits an object and is reflected. The receiving antenna 16a receives the reflected wave from the object. The specifying unit 121 synthesizes the modulated wave signal output from the transmission circuit 15 and the reflected wave signal output from the receiving circuit 16 to generate an intermediate frequency signal (hereinafter referred to as "IF signal"). The specifying unit 121 performs a fast Fourier transform (FFT) on the IF signal to obtain information on distance, velocity, and azimuth angle. The specifying unit 121 generates reflected wave data based on the obtained distance and azimuth angle information.

[0053] Each peak point in the reflected wave data is a reflection point. That is, the reflected wave data contains one or more reflection points. The identification unit 121 identifies the reflection points that reflect radio waves. More specifically, the identification unit 121 identifies the position of the reflection point, that is, the distance between the reflection point and the radio wave sensor 10, and the azimuth angle of the position where the reflection point exists. The position of the reflection point is expressed as a coordinate value in polar coordinates.

[0054] The identification unit 121 extracts reflection points, which are peak points included in the reflected wave data. The reflected wave data includes data showing the waveform of the reflected wave with respect to distance and data showing the waveform of the reflected wave with respect to angle. The identification unit 121 extracts peak points from both the waveform of the reflected wave with respect to distance and the waveform of the reflected wave with respect to angle. The identification unit 121 identifies the reflection points by associating the peak points in the reflected wave with respect to distance with the peak points in the reflected wave with respect to angle.

[0055] The detection unit 122 detects objects in the detection area 300 based on the reflection points identified by the identification unit 121.

[0056] The radio waves emitted from the radio wave sensor 10 may be reflected simultaneously by multiple objects. The detection unit 122 clusters the reflection points on the same object. For example, clustering is performed based on the distance between each reflection point. That is, the detection unit 122 includes multiple points within a certain distance range into one cluster (group of reflection points). For example, in addition to the distance between reflection points, clustering may also use the signal-to-noise ratio of each reflection point. In a specific example, the detection unit 122 determines a representative value for the reflection points belonging to the same cluster and sets the determined representative value to the position of a specific point (object). For example, the representative value is the centroid. However, the position of the object may be a representative value other than the centroid of the multiple reflection points. For example, the representative value may be the average value of the reflection points or the median value of the reflection points.

[0057] A transmitting signal (chirp) is transmitted from the transmitting antenna 15a at regular intervals. A moving object reflects the transmitting signal at two different points, and the receiving antenna 16a receives each reflected wave. The detection unit 122 detects the line-of-sight velocity of the object from the phase difference of multiple IF signals obtained from the moving object.

[0058] The tracking unit 123 tracks the detected specific point (object). Specifically, the tracking unit 123 assigns an ID to each specific point detected by the detection unit 122. The detection unit 122 outputs the detection results of the object's position and line-of-sight velocity at regular time intervals. The tracking unit 123 identifies the specific point that is the same as the specific point that was detected last time from among the specific points detected this time. For example, the tracking unit 123 estimates the current position of specific point a based on the direction and velocity of movement of specific point a from the previous time. The tracking unit 123 identifies the specific point that is closest to the position estimated from the previous direction and velocity of movement of specific point a as specific point a. Hereinafter, a specific point that has been assigned an ID and is being tracked will also be called a "target". A specific point identified as the same as a specific point that was detected last time will inherit the ID of the specific point that was detected last time.

[0059] The generation unit 124 generates confidence data. "Confidence data" is data that represents the degree of certainty that the target is a pedestrian crossing a crosswalk (hereinafter referred to as "pedestrian crossing") using predetermined identification information. In this embodiment, for example, a numerical value is used as the identification information for certainty, with confidence data = 1 meaning high certainty, and confidence data = 0 meaning low certainty or none. Certainty may be represented by three or more values ​​instead of two.

[0060] In the following, the above degree of certainty may be referred to as "reliability." In this case, a high degree of certainty means "reliable," and a low degree of certainty means "unreliable."

[0061] The detection results of the radio wave sensor 10 include "target data." Target data is data that represents a specific point that can be considered as the location of an object being tracked, as a point concept. In this embodiment, for example, it is determined whether a specific point that is the peak point of the reflected wave can be reliably identified as a pedestrian, so the target data may consist of the following information, for example.

[0062] Information 1: Location information of a specific point (e.g., XY coordinate values) Information 2: Speed ​​of movement of a specific point on the road (e.g., "m / s") Information 3: Identification information of a specific point (e.g., identification number) Information 4: Specific time of the specific point, i.e., time of ID assignment (e.g., absolute time or relative time) Information 5: Confidence data of the specific point (e.g., "1" or "0") However, since speed can be calculated by dividing the difference in position by the difference in time, the target data may be data with either Information 2 or Information 4 omitted.

[0063] The generation unit 124 executes, for example, one of the first generation process, second generation process, and third generation process described below. For example, the administrator (user) of the radio wave sensor 10 can activate one of the first generation process, second generation process, and third generation process and disable the other two by operating the radio wave sensor 10 or the setting support device 200 described later.

[0064] [First Generation Process for Confidence Data] Figure 6 is a flowchart of an example of the first generation process for confidence data. Hereinafter, the first generation process for confidence data will be abbreviated as "generation process 1". The meanings of the reference symbols and parameters in Figure 6 are as follows: TG: A specific point (target) that has been assigned an ID and is the target of tracking. CD: The crossing distance of target TG. The crossing distance CD is the distance traveled in the Y direction from the start of tracking of target TG (for example, t0 in the figure). ThD: A distance threshold for determining accuracy. Here, for example, it is set to "3m". RD: Confidence data. In this embodiment, RD=1 indicates reliability, and RD=0 indicates unreliability. AR: Pedestrian crossing area. The crossing area AR is, for example, zebra area 300_1.

[0065] (Contents of generation process 1) As shown in Figure 6, the processor 101 (generation unit 124) monitors whether the target TG is located within the range of the crossing area AR (step S101), and if it is within the range, calculates the current crossing distance CD (S102).

[0066] For example, if we define the starting point of tracking the target TG as "t0" and the current time as "t1", the difference between the Y coordinate value at time t0 and the Y coordinate value at time t1 is the crossing distance CD. Here, the Y axis extends in the longitudinal direction of the crossing area (i.e., the direction in which pedestrians cross the crosswalk 20). Therefore, the crossing distance CD is the distance component of the target (presumably a pedestrian) in the expected direction of movement (crossing direction).

[0067] Next, the processor 101 determines whether the inequality "CD ≥ ThD" holds true (S103). If the above inequality holds true, the processor 101 determines the value of the confidence data RD to be "1" (S104), and if the above inequality does not hold true, it determines the value of the confidence data RD to be "0" (S105).

[0068] (Tracking Example 1) The lower part of Figure 6 shows Tracking Example 1 of Target TG. In Tracking Example 1 of Figure 6, time t0 is the starting point of tracking Target TG, and the position of Target TG at each time point progresses in the order of time points t0 → t1 → t2 → t3 → t4.

[0069] In the example in Figure 6, the target TG with "RD = 0" is shown as a black circle, and the target TG with "RD = 1" is shown as a white circle. Also, in the example in Figure 6, the movement of the target TG is shown with arrows. That is, the tip of the arrow represents the direction of movement of the target TG, and the length of the arrow represents the speed of movement of the target TG.

[0070] (Temporal changes in tracking example 1) In tracking example 1 in Figure 6, at time t1, the transverse distance CD is 1m (<ThD), so RD is determined to be 0 and the target TG is shown as a black circle. At time t2, the transverse distance CD becomes 3m (≧ThD), so RD is determined to be 1 and the target TG is shown as a white circle. At times t3 and t4, the state of CD≧ThD continues, so RD is determined to be 1 and the target TG is shown as a white circle.

[0071] [Second Generation Process for Confidence Data] Figure 7 is a flowchart showing an example of the second generation process for confidence data. Hereafter, the second generation process for confidence data will be abbreviated as "generation process 2". The meanings of the reference numerals and parameters in Figure 7 (excluding those already defined in Figure 6) are as follows.

[0072] CT: This is the crossing time of the target TG. The crossing time CT is the elapsed time from the start of tracking the target TG (e.g., t0 in the diagram). ThT: This is the time threshold for accuracy determination. Here, for example, it is set to 3 seconds.

[0073] (Contents of generation process 2) As shown in Figure 7, the processor 101 (generation unit 124) of the radio wave sensor 10 monitors whether the target TG is located within the range of the crossing area AR (step S201), and if it is within the range, calculates the current crossing distance CD and crossing time CT (S202). For example, if the start time of tracking the target TG is "t0" and the current time is "t1", the difference between the Y coordinate value at time t0 and the Y coordinate value at time t1 becomes the crossing distance CD, and the time value (t1-t0) becomes the crossing time CT.

[0074] Next, the processor 101 determines whether both the inequalities "CD≧ThD" and "CT≧ThT" are true (S203). If both of the above inequalities are true, the processor 101 determines the value of the confidence data RD to be "1" (S204), and if neither of the above inequalities is true, it determines the value of the confidence data RD to be "0" (S205).

[0075] (Tracking Example 2) The lower part of Figure 7 shows Tracking Example 2 of the target TG. In Tracking Example 2 of Figure 7, the time intervals t1 to t4 are assumed to be 1 second, 2 seconds, 3 seconds, and 4 seconds, respectively.

[0076] (Temporal changes in tracking example 2) In tracking example 2 in Figure 7, at time t1, the crossing distance CD is 1 m (< ThD), so RD is determined to be 0 and the target TG is shown as a black circle. At time t2, the crossing distance CD becomes 3 m (≧ ThD), but the crossing time T is 2 seconds (< ThT), so RD is determined to be 0 and the target TG is shown as a black circle. At time t3, the crossing distance CD becomes 4 m (≧ ThD) and the crossing time CT becomes 3 seconds (≧ ThT), so RD is determined to be 1 and the target TG is shown as a white circle. At time t4, the state of CD≧ThD and CT≧ThT continues, so RD is determined to be 1 and the target TG is shown as a white circle.

[0077] [Third Generation Process for Confidence Data] Figure 8 is a flowchart showing an example of the third generation process for confidence data. Hereafter, the third generation process for confidence data will be abbreviated as "generation process 3". The meanings of the reference numerals and parameters in Figure 8 (excluding those already defined in Figure 6) are as follows.

[0078] VS: A vector representing the direction of movement of the target TG. The length of vector VS is arbitrary, but here it is assumed to be a unit vector. VY: A vector representing the Y-axis direction. The length of vector VY is arbitrary, but here it is assumed to be a unit vector. ThD: The dot product threshold for accuracy determination.

[0079] (Contents of generation process 3) As shown in Figure 8, the radio wave sensor 10 processor 101 (generation unit 124) monitors whether the target TG being tracked is located within the range of the crossing area AR (step S301), and if it is within the range, it calculates the current ABS (VS・VY) (S302). Note that "ABS()" is a function that takes the absolute value of the number in parentheses.

[0080] Next, the processor 101 determines whether the inequality ABS(VS・VY) ≤ ThI holds true (S303). If the above inequality holds true, the processor 101 determines the value of the confidence data RD to be "1" (S304), and if the above inequality does not hold true, it determines the value of the confidence data RD to be "0" (S305).

[0081] (Tracking Example 3) The lower part of Figure 8 shows Tracking Example 3 of target TG. In Tracking Example 3, at time points t1 and t2, the intersection angle between vectors VS and VY is large, so ABS(VS・VY) > ThI is assumed to hold. Furthermore, at time point t3, the intersection angle between vectors VS and VY becomes small, and ABS(VS・VY) ≤ ThI is assumed to hold.

[0082] (Temporal changes in tracking example 3) In tracking example 3 in Figure 8, at time points t1 and t2, ABS(VS・VY) > ThI, so RD = 0 and the target TG is shown as a black circle. At time point t3, ABS(VS・VY) ≤ ThI, so RD = 1 and the target TG is shown as a white circle. At time point t4, the state of ABS(VS・VY) ≤ ThI continues, so RD = 1 and the target TG is shown as a white circle.

[0083] The output unit 125 outputs the object detection results from the radio wave sensor 10. The detection results include the object's position (distance and azimuth), velocity, ID, time information indicating the detection time, and confidence data. Specifically, the detection results include the target data described above. The output unit 125 outputs the object detection results at regular time intervals.

[0084] For example, the generation unit 124 may store the object detection result from the radio wave sensor 10 in the non-volatile memory 102, and the output unit 125 may output the object detection result stored in the non-volatile memory 102.

[0085] [1-5. Setting Support Device] The radio wave sensor 10 is set using the setting support device 200. Specifically, the setting support device is used to set the detection area 300 of the radio wave sensor 10. For example, after the administrator (user) attaches the radio wave sensor 10 to the structure 50, the administrator connects the setting support device 200 to the radio wave sensor 10 and sets the radio wave sensor 10. Note that the configuration of the radio wave sensor 10 is the same as the configuration of the radio wave sensor 10 described in the first embodiment, so the same reference numerals are used for the same components and their descriptions are omitted.

[0086] [1-6. Hardware Configuration of the Setting Support Device] Figure 9 is a block diagram showing an example of the hardware configuration of the setting support device according to the first embodiment. The setting support device 200 according to the first embodiment includes a processor 201, a non-volatile memory 202, a volatile memory 203, an input / output interface (I / O) 204, a graphics controller 205, and a communication interface (communication I / F) 206. The setting support device 200 further includes an input device 211 and a display device 212. At least one of the input device 211 and the display device 212 may be an external device connected to the setting support device 200.

[0087] The volatile memory 203 is, for example, a semiconductor memory such as SRAM or DRAM. The non-volatile memory 202 is, for example, flash memory, a hard disk, or ROM. The non-volatile memory 202 stores the setting support program 210, which is a computer program, and the data used to execute the setting support program 210. Each function of the setting support device 200 is performed when the setting support program 210 is executed by the processor 201. The setting support program 210 can be stored in a recording medium such as flash memory, ROM, or CD-ROM. The processor 201 assists the administrator in setting up the radio wave sensor 10 using the setting support program 210.

[0088] The processor 201 is, for example, a CPU. However, the processor 201 is not limited to a CPU, and may be, for example, a GPU, or an ASIC, or a programmable logic device such as an FPGA or CPLD. In this case, the ASIC or programmable logic device is configured to perform the same processing as the setting support program 210.

[0089] For example, the input device 211 includes a keyboard and a pointing device such as a mouse. The input device 211 may also be a capacitive or pressure-sensitive touchpad superimposed on the screen of the display device 212. The input device 211 is used to input data to the setting support device 200. The input / output interface 204 is connected to the input device 211. The input / output interface 204 receives input data from the input device 211 and provides the received data to the processor 201.

[0090] The display device 212 includes, for example, a liquid crystal panel or an OEL (organic electroluminescent) panel. The display device 212 can display character or graphic information. The graphics controller 205 is connected to the display device 212 and controls the display on the display device 212. The graphics controller 205 includes, for example, a GPU and VRAM (Video RAM), holds data to be displayed on the display device 212 in VRAM, periodically reads one frame of video data from VRAM, and generates a video signal. The generated video signal is output to the display device 212, and the video is displayed on the display device 212. The functions of the graphics controller 205 may be included in the processor 201. A portion of the volatile memory 203 may be used as VRAM.

[0091] The communication interface 206 can communicate with external devices. For example, the communication interface 206 can be connected to the radio wave sensor 10 by a communication cable and communicate with the radio wave sensor 10. The communication interface 206 is a wireless communication interface and may also communicate with the radio wave sensor 10 wirelessly.

[0092] [1-7. Functions of the Setting Support Device] Figure 10 is a functional block diagram showing an example of the functions of the setting support device according to the first embodiment. When the processor 201 executes the setting support program 210, the setting support device 200 functions as an acquisition unit 221, a specification unit 222, a generation unit 223, a first determination unit 224, a first estimation unit 225, and a first output unit 226.

[0093] The radio wave sensor 10 can be set to two operating modes: a setting mode and an operation mode. The setting mode is the operating mode for setting up the radio wave sensor 10. The operation mode is the operating mode for detecting objects for traffic monitoring.

[0094] In setting mode, for example, a detection area 300 is set for the radio wave sensor 10. After the detection area 300 is set for the radio wave sensor 10, it is necessary to diagnose whether the set detection area 300 is appropriate. The setting support device 200 can diagnose whether the settings for the radio wave sensor 10 are appropriate.

[0095] The acquisition unit 221 acquires the object detection results from the radio wave sensor 10. In a specific example, the radio wave sensor 10 detects objects on the road during a certain detection period. For example, the detection period is a period of 1 second to 1 hour. For example, the detection period is a period that includes multiple control cycles. The radio wave sensor 10 continuously detects the position of objects during the detection period. The detection results output from the radio wave sensor 10 include target data for all objects detected during the detection period.

[0096] The identification unit 222 identifies low-reliability results from the detection results of objects (multiple objects) acquired by the acquisition unit 221. Low-reliability results are detection results in which the reliability data is lower than the standard throughout the entire period from when the detection of an object by the radio wave sensor 10 starts until when the detection of an object ends. Here, "start of detection" means the time when an object is first detected within the detection area 300, and "end of detection" means the time when the last object is detected within the detection area 300. The "standard" is RD = 1. That is, for example, if the reliability data for object a remains at 0 from the time object a detected by the radio wave sensor 10 enters the detection area 300 (e.g., zebra area 300_1) until object a exits the detection area 300 (zebra area 300_1), the detection result for object a is a low-reliability result. For example, if the confidence data changes from 0 to 1 during the period from when object b, detected by the radio wave sensor 10, enters the detection area 300 (zebra area 300_1) until object a exits the detection area 300 (zebra area 300_1), the detection result for object b is not a low-reliability result. A low-reliability result may be a detection result where the confidence data is lower than the standard at any of the following points: after the start of detection (a predetermined period from the start of detection), immediately before the end of detection (a predetermined period before the end of detection), or when a pedestrian passes a predetermined line in the detection area 300.

[0097] The identification unit 222 checks the confidence data included in each target data and identifies, for example, target data for which the confidence data remains 0 from the start of detection to the end of detection.

[0098] Figure 11 shows a first example of an inappropriate detection area setting. The zebra area 30_1 of the radio wave sensor 10 for detecting pedestrians crossing the road is set to cover the crosswalk 20, and to be the same size as or slightly larger than the crosswalk 20. By setting the zebra area 30_1 in this way, pedestrians can be accurately detected. In the example shown in Figure 11, the zebra area 30_1 is too large, and a part of the roadway 65 (intersection 61) adjacent to the crosswalk 20 is included in the zebra area 30_1. As a result, a vehicle V1 traveling on the roadway 65 is detected as a pedestrian crossing the road in the zebra area 30_1.

[0099] For example, if the generation process 2 described above is adopted, the crossing time CT of vehicle V1 does not exceed ThT (CT < ThT). Therefore, RD = 0 continues from the start to the end of detection of vehicle V1, and the target data for vehicle V1 will be a low-reliability result.

[0100] Figure 12 shows a second example of an inappropriate detection area setting. In the example shown in Figure 12, the zebra area 30_1 is too small relative to the crosswalk 20, and part of the crosswalk 20 is outside the zebra area 30_1. Therefore, if a pedestrian Pd1 is moving erratically across the crosswalk 20, the pedestrian Pd1 will repeatedly enter and exit the zebra area 30_1. Consequently, the radio wave sensor 10 repeatedly detects the pedestrian Pd1 for short periods of time.

[0101] For example, if the generation process 1 described above is adopted, the crossing distance CD of pedestrian Pd1 will not exceed ThD in any of the multiple detections (CD < ThD). Therefore, RD = 0 will continue from the start to the end of detection of pedestrian Pd1, and the target data for pedestrian Pd1 will be a low-reliability result.

[0102] Figure 13 shows a third example of an inappropriate detection area setting. In the example shown in Figure 13, the zebra area 30_1 is set at an inclination with respect to the crosswalk 20. Therefore, a pedestrian Pd2 moving across the crosswalk 20 is detected as moving in a direction inclined with respect to the longitudinal direction (Y-axis direction) of the zebra area 30_1.

[0103] For example, if the generation process 3 described above is adopted, the intersection angle between vectors VS and VY is large, and ABS(VS・VY) exceeds ThI (ABS(VS・VY) > ThI). As a result, RD = 0 continues from the start to the end of pedestrian Pd2 detection, and the target data for pedestrian Pd2 will be a low-reliability result.

[0104] Returning to Figure 10, the generation unit 223 generates low-reliability information regarding the occurrence of low-reliability results based on the low-reliability results identified by the identification unit 222.

[0105] (Low Reliability Information 1) The first example of low reliability information (hereinafter also referred to as "low reliability information 1") is the number of occurrences per unit time of detection results (target data) with low confidence data. Low reliability information 1 (Tlow1) is expressed by the following equation (1): Tlow1 = TDlow / DP ... (1) Here, TDlow is the number of detection results with low confidence data, and DP is the detection period. Tlow1 indicates the frequency of occurrence of target data with low confidence data (i.e., RD ​​= 0).

[0106] (Low Reliability Information 2) The second example of low reliability information (hereinafter also referred to as "low reliability information 2") is the ratio of the number of detection results (target data) with low confidence data to the total number of object detection results (target data). Low reliability information 2 (Tlow2) is expressed by the following equation (2): Tlow2 = TDlow / NTD ... (2) Here, NTD is the total number of target data. Tlow2 indicates the occurrence rate of target data with low confidence data (i.e., RD ​​= 0).

[0107] The first determination unit 224 determines whether the settings of the radio wave sensor 10 are appropriate based on the low reliability information generated by the generation unit 223. In a specific example, the first determination unit 224 compares the low reliability information with a threshold and determines that the settings of the radio wave sensor 10 are inappropriate if the low reliability information is greater than the threshold. The first determination unit 224 determines that the settings of the radio wave sensor 10 are appropriate if the low reliability information is less than or equal to the threshold.

[0108] (Appropriateness Determination 1) The first example of determining the appropriateness of the settings of the radio wave sensor 10 (hereinafter also referred to as "Appropriateness Determination 1") is a comparison of low reliability information 1 (Tlow1) and threshold 1 (ThA1). That is, in Appropriateness Determination 1, the first determination unit 224 compares Tlow1 and ThA1, and if Tlow1 is greater than ThA1 (Tlow1 > ThA1), it determines that the setting of the detection area 300 in the radio wave sensor 10 is inappropriate.

[0109] (Appropriateness Determination 2) The second example of determining the appropriateness of the settings of the radio wave sensor 10 (hereinafter also referred to as "Appropriateness Determination 2") is a comparison of low reliability information 2 (Tlow2) and threshold 2 (ThA2). That is, in Appropriateness Determination 2, the first determination unit 224 compares Tlow2 and ThA2, and if Tlow2 is greater than ThA2 (Tlow2 > ThA2), it determines that the setting of the detection area 300 in the radio wave sensor 10 is inappropriate.

[0110] For example, the first determination unit 224 selects either the suitability determination 1 or 2 described above. In a specific example, the administrator can operate the input device 211 to select either suitability determination 1 or 2 as the suitability determination of the settings of the radio wave sensor 10 performed by the first determination unit 224.

[0111] The first estimation unit 225 estimates the cause of the inappropriate setting of the radio wave sensor 10 when the first determination unit 224 determines that the setting of the radio wave sensor 10 is inappropriate.

[0112] For example, the first estimation unit 225 estimates the reason why the setting of the detection area 300 in the radio wave sensor 10 is inappropriate (hereinafter also referred to as "inappropriate cause"). Specifically, the following can be cited as inappropriate causes: Inappropriate cause 1: The set detection area 300 is too large. Inappropriate cause 2: The set detection area 300 is too small. Inappropriate cause 3: The set detection area 300 is tilted.

[0113] In a specific example, the first estimation unit 225 estimates the cause of the inappropriateness based on the location where the low-reliability result occurred. For example, as shown in Figure 11, if the size of the detection area 30 is too large, many vehicles traveling on the roadway 65 will pass the edge of the detection area 30 (the right edge in Figure 11). Therefore, low-reliability results will be concentrated at the edge of the detection area 300. When low-reliability results are concentrated at the edge of the detection area 300, the first estimation unit 225 estimates that the reason why the setting of the detection area 300 is inappropriate is the cause of inappropriateness 1.

[0114] For example, as shown in Figure 12, if the size of the detection area 30 is too small, the pedestrian Pd1 will be detected intermittently along one side of the detection area 300 (the right side in Figure 11). Therefore, the first estimation unit 225 estimates that the reason why the detection area 300 is not set appropriately is due to inappropriate cause 2 when low-reliability results are concentrated along one side of the detection area 300.

[0115] For example, as shown in Figure 13, if the detection area 30 is inclined with respect to the crosswalk 20, the pedestrian Pd2 will move along a direction inclined with respect to the longitudinal direction (Y direction) of the detection area 300. Therefore, if a large number of low-reliability results occur in which the pedestrian moves along a direction inclined with respect to the longitudinal direction of the detection area 300, the first estimation unit 225 estimates that the reason why the detection area 300 is not set appropriately is due to inappropriate cause 3.

[0116] Returning to Figure 10, the first output unit 226 outputs the result of the determination made by the first determination unit 224. In a specific example, the first output unit 226 displays on the display device 212 the determination result of whether or not the setting of the detection area 300 is appropriate, as determined by the first determination unit 224.

[0117] Furthermore, the first output unit 226 outputs the inappropriate cause estimated by the first estimation unit 225. That is, the first output unit 226 displays on the display device 212 the estimated result of which of the inappropriate causes 1 to 3 is the reason why the detection area 300 is not set properly.

[0118] Figure 14 shows an example of the screen display when the detection area setting is incorrect (for example, when the cause of the incorrect setting is incorrect cause 1). For example, the screen of the display device 212 displays the words, "The detection area setting is incorrect. Please reset the detection area," and "The size of the detection area is too large." Furthermore, the screen of the display device 212 displays a diagram indicating that the detection area 300 is too large.

[0119] [1-8. Operation of the setting support device] Figure 15 is a flowchart showing an example of the setting diagnostic process for a radio wave sensor by the setting support device according to the first embodiment.

[0120] After the detection area 300 of the radio wave sensor 10 is set, the radio wave sensor 10 detects objects during the detection period for setting diagnostic purposes. The object detection results are stored in the non-volatile memory 102 of the radio wave sensor 10.

[0121] When performing a configuration diagnosis, the administrator of the radio wave sensor 10 connects the configuration support device 200 to the radio wave sensor 10. The radio wave sensor 10 transmits the detection results stored in the non-volatile memory 102. The processor 201 of the configuration support device 200 acquires the detection results transmitted from the radio wave sensor 10 (step S111).

[0122] The processor 201 identifies low-reliability results from the acquired detection results (S112). Specifically, the processor 201 refers to the target data included in the acquired detection results and identifies target data (ID) for which the reliability data is lower than the standard (RD = 0) from the start to the end of detection.

[0123] The processor 201 generates low reliability information (S113). Specifically, the processor 201 calculates either the low reliability information 1 or the low reliability information 2 described above. For example, the processor 201 counts the low reliability results included in the detection results (the number of occurrences of low reliability results during the detection period) and calculates low reliability information 1 by dividing the number of occurrences of low reliability results by the detection period. In another example, the processor 201 counts the total number of acquired detection results and the low reliability results included in the detection results (the number of occurrences of low reliability results during the detection period), and calculates low reliability information 2 by dividing the number of occurrences of low reliability results by the total number of detection results.

[0124] The processor 201 determines whether the settings of the radio wave sensor 10 are appropriate by comparing the low reliability information with a threshold (S114). If suitability determination 1 is selected, the processor 201 compares Tlow1 and ThA1, and if Tlow1 > ThA1, it determines that the setting of the detection area 300 of the radio wave sensor 10 is inappropriate, and if Tlow1 ≤ ThA1, it determines that the setting of the detection area 300 of the radio wave sensor 10 is appropriate. If suitability determination 2 is selected, the processor 201 compares Tlow2 and ThA2, and if Tlow2 > ThA2, it determines that the setting of the detection area 300 of the radio wave sensor 10 is inappropriate, and if Tlow2 ≤ ThA2, it determines that the setting of the detection area 300 of the radio wave sensor 10 is appropriate.

[0125] If the processor determines that the detection area 300 of the radio wave sensor 10 is set appropriately (in S114, "appropriate"), the processor 201 displays information (for example, text information) indicating that the detection area 300 of the radio wave sensor 10 is set appropriately on the display device 212 (S115). This completes the setting diagnostic process.

[0126] If the processor determines that the setting of the detection area 300 of the radio wave sensor 10 is inappropriate (indicated as "inappropriate" in S114), the processor 201 estimates the reason why the setting of the detection area 300 is inappropriate (S116). Specifically, the processor 201 estimates which of the inappropriate causes 1 to 3 is the reason for the inappropriate setting of the detection area 300, based on the location where the low reliability result occurs.

[0127] The processor 201 displays information on the display device 212 indicating that the setting of the detection area 300 of the radio wave sensor 10 is not appropriate, and the reason why the setting of the detection area 300 of the radio wave sensor 10 is not appropriate (S117). This completes the setting diagnostic process.

[0128] [2. Second Embodiment] The setting support device 200 according to the second embodiment diagnoses abnormalities in the radio wave sensor 10. The hardware configuration of the setting support device 200 according to the second embodiment is the same as that of the setting support device 200 according to the first embodiment, so a description is omitted.

[0129] [2-1. Functions of the Setting Support Device] Figure 16 is a functional block diagram showing an example of the functions of the setting support device according to the second embodiment. When the processor 201 executes the setting support program 210, the setting support device 200 functions as an acquisition unit 221, a specification unit 222, a generation unit 223, a second determination unit 227, a second estimation unit 228, and a second output unit 229.

[0130] The basic functions of the acquisition unit 221, the identification unit 222, and the generation unit 223 according to the second embodiment are the same as those of the acquisition unit 221, the identification unit 222, and the generation unit 223 according to the first embodiment, so their explanation will be omitted.

[0131] The second determination unit 227 determines whether the radio wave sensor 10 is abnormal based on the low reliability information generated by the generation unit 223. In a specific example, the second determination unit 227 determines whether the radio wave sensor 10 is abnormal based on the difference between the first low reliability information and the second low reliability information. The first low reliability information is low reliability information generated by the generation unit 223 based on the object detection result of the radio wave sensor 10 at a first time point, and the second low reliability information is low reliability information generated by the generation unit 223 based on the object detection result of the radio wave sensor 10 at a second time point, which is after the first time point.

[0132] For example, the first point in time is the time when the detection area 300 of the radio wave sensor 10 is set, and the time when the radio wave sensor 10 is put into operation. For example, the second point in time is the time immediately before the abnormality diagnosis of the radio wave sensor 10 is performed, and the time when the operation of the radio wave sensor 10 is interrupted for the purpose of abnormality diagnosis.

[0133] For example, the detection area 300 of the radio wave sensor 10 is set, and at the start of operation of the radio wave sensor 10 (first time point), the radio wave sensor 10 is made to detect an object, and based on the detection result at the first time point, the generation unit 223 generates first low reliability information. For example, the first low reliability information is generated at the first time point of the radio wave sensor 10 and stored in the non-volatile memory 202 of the setting support device 200, the non-volatile memory 102 of the radio wave sensor 10, or other storage device. In another example, the object detection result at the first time point may be stored in the non-volatile memory 102 of the radio wave sensor 10, the non-volatile memory 202 of the setting support device 200, or other storage device, and at the time of abnormality diagnosis of the radio wave sensor 10, the generation unit 223 may generate first low reliability information based on the detection result read from the storage device.

[0134] For example, the first low-reliability information may be the low-reliability information used to determine the appropriateness of the settings of the radio wave sensor 10 as described in the first embodiment. In this case, the low-reliability information when it is determined that the settings of the radio wave sensor 10 are appropriate is used as the first low-reliability information.

[0135] When an abnormality diagnosis is initiated, the generation unit 223 generates first low-reliability information based on the object detection result at a first point in time stored in the storage device as described above, or reads first low-reliability information that has been previously generated and stored in the storage device from the storage device.

[0136] Furthermore, the generation unit 223 generates second low-reliability information based on the object detection result at a second point in time obtained from the radio wave sensor 10, which was interrupted in order to initiate abnormality diagnosis.

[0137] For example, the radio wave sensor 10 in operation continuously detects objects. For example, the non-volatile memory 102 stores the detection results of objects for the most recent fixed period (detection period). That is, new detection results are added to the detection results for the detection period stored in the non-volatile memory 102, and the oldest detection results are deleted. As a result, if the radio wave sensor 10 is stopped for abnormality diagnosis, the detection results for the last fixed period are stored in the non-volatile memory 102. The acquisition unit 221 acquires the detection results (detection results at a second point in time) stored in the non-volatile memory 102 of the radio wave sensor 10.

[0138] The second determination unit 227 calculates the difference (Diff) between the first low reliability information (Tlow_1) and the second low reliability information (Tlow_2) from equation (3). Diff = Tlow_2 - Tlow_1 ... (3) Here, the first low reliability information and the second low reliability information are each low reliability information 1 (frequency of occurrence of target data with low reliability data). However, the first low reliability information and the second low reliability information may also be low reliability information 2 (rate of occurrence of target data with low reliability data).

[0139] For example, if the detection area 300 of the radio wave sensor 10 is set appropriately at the first time point, and no abnormality occurs in the radio wave sensor 10, the frequency of low reliability results is low, and the first low reliability information (Tlow_1) is low. On the other hand, for example, if the orientation of the radio wave sensor 10 changes due to collision with an obstacle, strong winds, etc., during operation of the radio wave sensor 10, the detection area 30 may shift away from the pedestrian crossing 20. During operation of the radio wave sensor 10, the radio wave sensor 10 may deteriorate, causing the radio wave irradiation area of ​​the radio wave sensor 10 to become smaller than the detection area 30. When an abnormality occurs in the radio wave sensor 10 (change in orientation, deterioration, etc.), the frequency of low reliability results at the second time point increases, and the second low reliability information (Tlow_1) is high. Therefore, if no abnormality occurs in the radio wave sensor 10, the difference (Diff) is small, and if an abnormality occurs in the radio wave sensor 10, the difference (Diff) is large.

[0140] The second determination unit 227 compares the difference (Diff) with a threshold (ThB). If the difference is greater than the threshold (Diff > ThB), it determines that the radio wave sensor 10 is abnormal. If the difference is less than the threshold (Diff ≤ ThB), it determines that the radio wave sensor 10 is normal. The threshold ThB is a relative threshold used for comparison with the relative value difference (Diff).

[0141] The second estimation unit 228 estimates the cause of the abnormality in the radio wave sensor 10 (hereinafter also referred to as the "cause of the abnormality") when the second determination unit 227 determines that the radio wave sensor 10 is abnormal.

[0142] Specifically, the following are possible causes of the abnormality: Cause 1: The detection area 300 is misaligned with the pedestrian crossing 20. Cause 2: The radio wave irradiation area is smaller than the detection area 300. Cause 3: The detection area 300 is tilted.

[0143] In a specific example, the second estimation unit 228 estimates the cause of the anomaly based on the location where the low-reliability result occurred.

[0144] Figure 17 shows the first example of a malfunction in the radio wave sensor. In the example shown in Figure 17, the orientation of the radio wave sensor 10 changes due to strong winds, collision with an obstacle, etc., causing the zebra area 30_1 to shift away from the pedestrian crossing 20 (cause of malfunction 1), and a portion of the roadway 65 (intersection 61) adjacent to the pedestrian crossing 20 is included in the zebra area 30_1. As a result, a vehicle V2 traveling on the roadway 65 is detected as a pedestrian crossing in the zebra area 30_1.

[0145] For example, as shown in Figure 17, if the detection area 30 is offset from the pedestrian crossing 20, many vehicles traveling on the roadway 65 will pass the edge of the detection area 30 (the right edge in Figure 17). Therefore, low-reliability results will be concentrated at the edge of the detection area 300. The second estimation unit 228 estimates that the cause of the malfunction in the radio wave sensor 10 is malfunction cause 1 when low-reliability results are concentrated at the edge of the detection area 300.

[0146] Figure 18 shows a second example of a malfunction in the radio wave sensor. In the example shown in Figure 18, the zebra area 30_1 is appropriate for the pedestrian crossing 20, but the radio wave irradiation area 35 is smaller than the zebra area 30_1, and a part of the pedestrian crossing 20 is outside the radio wave irradiation area 35. Therefore, when pedestrian Pd3 is moving near the outer edge of the radio wave irradiation area 35, the radio wave sensor 10 repeatedly detects and fails to detect pedestrian Pd3.

[0147] For example, as shown in Figure 18, if the radio wave irradiation area 35 is smaller than the detection area 30, the pedestrian Pd3 crossing the street is intermittently detected within the zebra area 300_1. Therefore, the second estimation unit 228 estimates that the cause of the abnormality in the radio wave sensor 10 is abnormality cause 2 when low reliability results are concentrated within the zebra area 300_1.

[0148] Due to strong winds, collisions with obstacles, etc., the orientation of the radio wave sensor 10 may change, causing the zebra area 30_1 to tilt relative to the crosswalk 20, as shown in Figure 13. In this case, the pedestrian Pd2 will move along a direction that is tilted relative to the longitudinal direction (Y direction) of the detection area 300. Therefore, the second estimation unit 228 estimates that the cause of the malfunction in the radio wave sensor 10 is the malfunction cause 3 when a large number of low-reliability results occur where the pedestrian moves in a direction tilted relative to the longitudinal direction of the detection area 300.

[0149] Returning to Figure 16, the second output unit 229 outputs the result of the determination made by the second determination unit 227. In a specific example, the second output unit 229 displays the determination result of whether or not the radio wave sensor 10 is abnormal, as determined by the second determination unit 227, on the display device 212.

[0150] Furthermore, the second output unit 229 outputs the cause of the abnormality estimated by the second estimation unit 228. That is, the second output unit 229 displays on the display device 212 the estimated result of which of the abnormality causes 1 to 3 the cause of the abnormality in the radio wave sensor 10 is.

[0151] Figure 19 shows an example of a screen display when the radio wave sensor is malfunctioning. In the example in Figure 19, the cause of the malfunction in the radio wave sensor 10 is malfunction cause 1. For example, the screen of the display device 212 displays the words "Radio wave sensor is malfunctioning." and "Detection area is shifted from the pedestrian crossing." Furthermore, the screen of the display device 212 displays a diagram showing that the detection area 30 is shifted from the pedestrian crossing 20.

[0152] [2-2. Operation of the setting support device] Figure 20 is a flowchart showing an example of abnormality diagnosis processing of a radio wave sensor by the setting support device according to the second embodiment.

[0153] When performing an abnormality diagnosis on the radio wave sensor 10, the administrator interrupts the operation of the radio wave sensor 10, and the radio wave sensor 10 stops detecting objects. The non-volatile memory of the radio wave sensor 10 stores the object detection results (detection results at a second point in time) for the most recent period (detection period).

[0154] When performing an abnormality diagnosis, the administrator of the radio wave sensor 10 connects the setting support device 200 to the radio wave sensor 10. The radio wave sensor 10 transmits the detection results stored in the non-volatile memory 102. The processor 201 of the setting support device 200 acquires the detection results transmitted from the radio wave sensor 10 (step S211).

[0155] The non-volatile memory 202 of the setting support device 200 stores the first low reliability information used to determine the suitability of the settings during the initial setup of the radio wave sensor 10. The processor 201 reads the first low reliability information from the non-volatile memory 202 (S212).

[0156] The processor 201 identifies low-reliability results from the acquired detection results (S213). Furthermore, the processor 201 generates second low-reliability information (S214). Since S213 and S214 are the same as S112 and S113 in the first embodiment, their explanation is omitted.

[0157] The processor 201 calculates the difference (Diff) between the first low reliability information (Tlow_1) and the second low reliability information (Tlow_2) (S215). The processor 201 determines whether the radio wave sensor 10 is abnormal or not by comparing the difference (Diff) with the threshold (ThB) (S216).

[0158] If the radio wave sensor 10 is determined to be functioning normally (in S216, "normal"), the processor 201 displays information indicating that the radio wave sensor 10 is functioning normally (for example, text information) on the display device 212 (S217). This completes the abnormality diagnosis process.

[0159] If the radio wave sensor 10 is determined to be abnormal (in S216, "abnormal"), the processor 201 estimates the cause of the abnormality in the radio wave sensor 10 (S218). Specifically, the processor 201 estimates which of the abnormality causes 1 to 3 is the cause of the abnormality in the radio wave sensor 10, based on the location where the low reliability result occurs.

[0160] The processor 201 displays information indicating that the radio wave sensor 10 is malfunctioning, and the cause of the malfunction in the radio wave sensor 10, on the display device 212 (S219). This completes the malfunction diagnosis process.

[0161] [3. Third Embodiment] The setting support device 200 according to the third embodiment diagnoses abnormalities in the radio wave sensor 10 using second low reliability information (Tlow_2) instead of difference (Diff). In the third embodiment, the second low reliability information is simply referred to as "low reliability information".

[0162] Referring to Figure 16, the functions of the setting support device 200 according to the third embodiment will be described. Note that the functions of the acquisition unit 221, the identification unit 222, the generation unit 223, the second estimation unit 228, and the second output unit 229 are the same as those of the acquisition unit 221, the identification unit 222, the generation unit 223, the second estimation unit 228, and the second output unit 229 in the second embodiment, so their description will be omitted.

[0163] The second determination unit 227 determines whether the radio wave sensor 10 is abnormal based on the low reliability information generated by the generation unit 223. In a specific example, the second determination unit 227 compares the low reliability information with a threshold and determines that the radio wave sensor 10 is abnormal if the low reliability information is greater than the threshold. The second determination unit 227 determines that the radio wave sensor 10 is normal if the low reliability information is less than or equal to the threshold.

[0164] (Anomaly Determination 1) The first example of anomaly determination of the radio wave sensor 10 (hereinafter also referred to as "anomaly determination 1") is a comparison between low reliability information 1 (Tlow1) and threshold 1 (ThB1). That is, in anomaly determination 1, the second determination unit 227 compares Tlow1 and ThB1, and determines that the radio wave sensor 10 is abnormal if Tlow1 is greater than ThB1 (Tlow1 > ThB1). Threshold 1 (ThB1) is an absolute threshold used for comparison with low reliability information 1 (Tlow1), which is an absolute value.

[0165] (Anomaly Determination 2) The second example of anomaly determination of the radio wave sensor 10 (hereinafter also referred to as "anomaly determination 2") is a comparison between low reliability information 2 (Tlow2) and threshold 2 (ThB2). That is, in anomaly determination 2, the second determination unit 227 compares Tlow2 and ThB2, and determines that the radio wave sensor 10 is abnormal if Tlow2 is greater than ThB2 (Tlow2 > ThB2). Threshold 2 (ThB2) is an absolute threshold used for comparison with low reliability information 2 (Tlow2), which is an absolute value.

[0166] For example, the first determination unit 224 selects either the suitability determination 1 or 2 described above. In a specific example, the administrator can operate the input device 211 to select either suitability determination 1 or 2 as the suitability determination of the settings of the radio wave sensor 10 performed by the first determination unit 224.

[0167] The second determination unit 227 may have both the functions of the second determination unit 227 in the second embodiment and the functions of the second determination unit 227 in the third embodiment.

[0168] [4. Fourth Embodiment] The setting support device 200 according to the fourth embodiment sets an exclusion area (hereinafter also referred to as the "mask area") in the detection area 300 that is excluded from the detection target of the radio wave sensor 10. The hardware configuration of the setting support device 200 according to the fourth embodiment is the same as the hardware configuration of the setting support device 200 according to the first embodiment, so a description is omitted.

[0169] [4-1. Functions of the Setting Support Device] Figure 21 is a functional block diagram showing an example of the functions of the setting support device according to the fourth embodiment. When the processor 201 executes the setting support program 210, the setting support device 200 functions as an acquisition unit 221, a specification unit 222, a first determination unit 230, a second determination unit 231, and a third output unit 232. The functions of the acquisition unit 221 and the specification unit 222 are the same as those of the acquisition unit 221 and the specification unit 222 in the first embodiment, so their explanation is omitted.

[0170] The first determination unit 230 determines candidate areas that are candidates for the mask area based on the detection position of the object in the low-reliability result identified by the identification unit 222.

[0171] On roads such as pedestrian crossings, there are locations where low-reliability results frequently occur, depending on the road's configuration (positional relationship to intersections, number of lanes, etc.) and the placement of fixed objects such as traffic lights, signs, trees, and guardrails. In order to prevent false detection of objects other than the target object, the setting support device 200 according to the fourth embodiment sets a mask area at locations where low-reliability results frequently occur.

[0172] Figure 22 shows an example of a location where low-reliability results frequently occur. For example, at a pedestrian crossing 20 adjacent to an intersection 61, vehicles V3 and V4 turning left or right from a roadway 65 running parallel to the pedestrian crossing 20 onto the roadway 60 where the pedestrian crossing 20 is located may enter the pedestrian crossing 20 at an angle (i.e., not in a direction perpendicular to the pedestrian crossing 20). Therefore, the entry vectors of vehicles V3 and V4 into the pedestrian crossing 20 include a longitudinal (crossing) component of the pedestrian crossing 20 and a widthwise component of the pedestrian crossing 20. Consequently, the radio wave sensor 10 detects vehicles V3 and V4 as pedestrians crossing the road.

[0173] In Figure 22, detection result 71 is the result of detecting vehicle V3 turning left from roadway 65 to roadway 60, and detection result 72 is the result of detecting vehicle V4 turning right from roadway 65 to roadway 60. From the time vehicles V3 and V4 enter the detection area 30 until they exit the detection area 30, they move roughly in the width direction of the detection area 30. Therefore, if generation process 1 or generation process 3 described above is selected, RD = 0 will continue from the start to the end of detection, and the target data for vehicles V3 and V4 will be a low-reliability result. Since vehicles V3 and V4 move at a high speed and pass through the detection area 30 in a short time, even if generation process 2 is selected, RD = 0 will continue from the start to the end of detection, and the target data for vehicles V3 and V4 will be a low-reliability result.

[0174] Multiple vehicles turning left from roadway 65 onto roadway 60 generally follow the same route and are detected as pedestrians at approximately the same location on the crosswalk 20. Similarly, multiple vehicles turning right from roadway 65 onto roadway 60 are detected as pedestrians at approximately the same location on the crosswalk 20. Therefore, vehicles turning left or right from roadway 65 onto roadway 60 appear frequently in a certain area as a low-reliability result.

[0175] For example, trees are fixedly positioned near roads, but their leaves and branches sway in the wind, causing them to be detected as moving pedestrians. Therefore, trees appear frequently in certain areas as a result of low reliability.

[0176] In Figure 22, detection result 73 indicates that a tree TR near the pedestrian crossing 20 was detected. Because a tree TR has numerous leaves and branches, it is detected as multiple pedestrians crossing the road, concentrated in a certain area. Furthermore, because the leaves and branches sway in various directions due to the wind, the detected movement distance is short, and the movement vector changes in various ways over time. Therefore, if generation process 1 or generation process 3 described above is selected, RD = 0 will continue from the start to the end of detection, resulting in low-reliability results for the target data of the tree TR. Moreover, the movement of leaves and branches is irregular, and even if they are detected as a single object, the detection time is short. Therefore, if generation process 2 is selected, RD = 0 will continue from the start to the end of detection, resulting in low-reliability results for the target data of the tree TR.

[0177] For example, the first determination unit 230 clusters the target data of low-reliability results based on their location. Figure 23 shows an example of the clustering results of low-reliability results. As described above, objects other than the detection target (pedestrians crossing the road) (vehicles, trees) are concentrated in the same area and detected as multiple low-reliability results. Therefore, through clustering, for example, a group of vehicles turning left becomes one cluster CL1, a group of vehicles turning right becomes one cluster CL2, and one tree becomes one cluster CL3. The first determination unit 230 determines each cluster CL1, CL2, and CL3 as a candidate area.

[0178] Returning to Figure 21, the third output unit 232 outputs the candidate area determined by the first determination unit 230. In a specific example, the third output unit 232 displays the candidate area in the sensor coordinate space on the display device 212. More specifically, the third output unit 232 superimposes the detection area 300 and the candidate area on the display device 212.

[0179] The display device 212 displays a mask area setting screen in which candidate areas are placed in the sensor coordinate space. The third output unit 232 causes the candidate areas to be displayed on the display device 212 along with the detection area 300. That is, in the mask area setting screen, the detection area 300 and the candidate areas are displayed simultaneously in the sensor coordinate space defined by the X and Y axes.

[0180] By displaying the detection area 300 and the candidate areas simultaneously, the administrator can understand the positional relationship between each candidate area and the detection area 300. For example, by comparing the positional relationship between each candidate area and the detection area 300 with the positional relationship between each object in real space and the detection area 30 (or pedestrian crossing 20), the administrator can recognize the position of the candidate area relative to the pedestrian crossing 20.

[0181] The second decision unit 231 determines the candidate area displayed on the display device 212 as the mask area. For example, an administrator can specify a candidate area to be used as the mask area using the input device 211. The second decision unit 231 determines the candidate area specified by the administrator as the mask area.

[0182] The second determination unit 231 generates setting information 111 for setting the radio wave sensor 10. The setting information 111 includes mask area information, which specifies the position, size, and shape of the mask area.

[0183] For example, the setting support device 200 sets the radio wave sensor 10 by transmitting the generated setting information 111 to the radio wave sensor 10. The radio wave sensor 10 can set a mask area by storing the received setting information 111 in the non-volatile memory 102. In the operating mode, the radio wave sensor 10 with the mask area set excludes objects detected in the mask area from the detection results.

[0184] [4-2. Operation of the setting support device] Figure 24 is a flowchart showing an example of the setting support process for a radio wave sensor by the setting support device according to the fourth embodiment.

[0185] The processor 201 of the setting support device 200 starts the radio wave sensor 10 in setting mode (step S311).

[0186] When the radio wave sensor 10 is activated in setting mode, the radio wave sensor 10 detects an object (position and velocity) for a certain detection period. The radio wave sensor 10 transmits the detection result. The processor 201 acquires the detection result transmitted from the radio wave sensor 10 by receiving it (S312).

[0187] The processor 201 identifies the low-reliability result from the acquired detection results (S313). S313 is the same as S112 in the first embodiment, so its explanation is omitted.

[0188] The processor 201 clusters the low-reliability results based on their location (S314). Furthermore, the processor 201 determines candidate areas for the clusters of the low-reliability results (S315).

[0189] The processor 201 causes the display device 212 to display the detection area and the candidate area (S316).

[0190] The administrator specifies candidate areas to be designated as mask areas using the input device 211. The processor 201 accepts the designation of candidate areas to be designated as mask areas (S317).

[0191] The processor 201 determines the candidate area specified by the administrator to be the mask area (S318).

[0192] The processor 201 generates setting information 111 specifying the position, size, and shape of each determined mask area (S319). The processor 201 transmits the generated setting information 111 to the radio wave sensor 10, thereby setting the mask area on the radio wave sensor 10 (S320). This completes the setting support process.

[0193] [4-3. Modifications] In the fourth embodiment, candidate areas were determined based on low reliability results, and the candidate areas designated by the administrator were determined as mask areas, but the embodiment is not limited thereto. For example, the setting support device 200 may cluster the low reliability results and determine the cluster of low reliability results as mask areas.

[0194] [5. Fifth Embodiment] The radio wave sensor 10 according to the fifth embodiment identifies low reliability results from the object detection results during operation and self-diagnoses abnormalities in the radio wave sensor 10. The hardware configuration of the radio wave sensor 10 according to the fifth embodiment is the same as the hardware configuration of the radio wave sensor 10 according to the first embodiment, so a description is omitted.

[0195] [5-1. Functions of the Radio Wave Sensor] Figure 25 is a functional block diagram showing an example of the functions of the radio wave sensor according to the fifth embodiment. When the processor 101 executes the control program 110, the radio wave sensor 10 functions as a first identification unit 131, a detection unit 132, a tracking unit 133, a first generation unit 134, an acquisition unit 135, a second identification unit 136, a second generation unit 137, a determination unit 138, an estimation unit 139, and an output unit 140.

[0196] The functions of the first identification unit 131, the detection unit 132, the tracking unit 133, and the first generation unit 134 are the same as those of the identification unit 121, the detection unit 122, the tracking unit 123, and the generation unit 124 described in the first embodiment, so their explanation will be omitted.

[0197] For example, the self-diagnosis of the radio wave sensor 10 is performed periodically, such as weekly, monthly, or annually.

[0198] The acquisition unit 135 acquires detection results by reading the object detection results from the non-volatile memory 102. More specifically, when the timing for performing a self-diagnosis arrives, the acquisition unit 135 reads the latest detection results (target data) for the detection period from the non-volatile memory 102.

[0199] The functions of the second identification unit 136 and the second generation unit 137 are the same as those of the identification unit 222 and the generation unit 223 described in the first embodiment, so their description is omitted. The functions of the determination unit 138 and the estimation unit 139 are the same as those of the second determination unit 227 and the second estimation unit 228 described in the second embodiment.

[0200] In other words, the determination unit 138 determines whether or not the radio wave sensor 10 is abnormal based on the low reliability information generated by the second generation unit 137. In a specific example, the determination unit 138 determines whether or not the radio wave sensor 10 is abnormal based on the difference between the first low reliability information and the second low reliability information.

[0201] The estimation unit 139 estimates the cause of the abnormality in the radio wave sensor 10 when the determination unit 138 determines that the radio wave sensor 10 is abnormal. In a specific example, the estimation unit 139 estimates which of the abnormality causes 1 to 3 is the cause of the abnormality in the radio wave sensor 10, based on the location where the low-reliability result occurred.

[0202] The output unit 140 notifies an external device of the abnormality of the radio wave sensor 10 when the determination unit 138 determines that the radio wave sensor 10 is abnormal. The external device is, for example, a server operated by a road traffic information management organization, a terminal used by an administrator, or a setting support device 200. The output unit 140 may also notify the external device of the estimated result of the cause of the abnormality by the estimation unit 139. This allows for periodic abnormality diagnosis of the radio wave sensor 10, and if an abnormality is determined, the external device is quickly notified.

[0203] [5-2. Operation of the Radio Wave Sensor] The operation of the radio wave sensor 10 according to the fifth embodiment will be described below.

[0204] Figure 26 is a flowchart showing an example of an anomaly diagnosis process for a radio wave sensor according to the fifth embodiment.

[0205] When the processor 101 starts the control program 110, the radio wave sensor 10 starts operating. The processor 101 determines whether or not the timing for starting the abnormality diagnosis has arrived (step S401). If the timing for starting the abnormality diagnosis has not arrived (NO in S401), the processor 101 returns to S401.

[0206] When the timing for starting abnormality diagnosis arrives (YES in S401), the processor 101 reads the detection results for the detection period from the non-volatile memory 102 (S402).

[0207] The non-volatile memory 102 stores the first low reliability information for the initial setup of the radio wave sensor 10. The processor 201 reads the first low reliability information from the non-volatile memory 102 (S403).

[0208] The processor 101 identifies a low-reliability result from the read detection result (S404). Furthermore, the processor 101 generates second low-reliability information (S405). Steps S404 and S405 are the same as steps S112 and S113 in the first embodiment, so their explanation is omitted.

[0209] The processor 101 calculates the difference (Diff) between the first low reliability information (Tlow_1) and the second low reliability information (Tlow_2) (S406).

[0210] The processor 101 determines whether the radio wave sensor 10 is abnormal by comparing the difference (Diff) and the threshold (ThB) (S407).

[0211] If the radio wave sensor 10 is determined to be functioning normally (in S407, "normal"), the processor 101 transmits information indicating that the radio wave sensor 10 is functioning normally (for example, character information) to an external device, notifying that the radio wave sensor 10 is functioning normally (S408). This completes the abnormality diagnosis process.

[0212] If the radio wave sensor 10 is determined to be abnormal (in S407, "abnormal"), the processor 101 estimates the cause of the abnormality in the radio wave sensor 10 (S409). Specifically, the processor 101 estimates which of the abnormality causes 1 to 3 is the cause of the abnormality in the radio wave sensor 10, based on the location where the low reliability result occurs.

[0213] The processor 101 transmits information indicating that the radio wave sensor 10 is malfunctioning, as well as the cause of the malfunction in the radio wave sensor 10, to an external device, notifying it that the radio wave sensor 10 is malfunctioning (S410). This completes the malfunction diagnosis process.

[0214] [5-3. Modified Examples] In the fifth embodiment, the radio wave sensor 10 determines whether or not it is abnormal by comparing the difference (Diff) between the first low reliability information (Tlow_1) and the second low reliability information (Tlow_2) with a difference threshold (ThB), but is not limited to this. The abnormality of the radio wave sensor 10 may be diagnosed by comparing the second low reliability information (Tlow_2) with an absolute threshold (i.e., by a process similar to the abnormality diagnosis in the second embodiment).

[0215] [6. Other Modifications] In the embodiments described above, the confidence data was set to binary data ("1" or "0"), but it is not limited to this. The confidence data may also be real number data. For example, in generation process 1, RD = 0 when the crossing distance CD is 0, and RD = 1 when the crossing distance CD is ThD or greater, and RD can be linearly changed in the range from 0 to 1 according to the value of the crossing distance CD in the range from 0 to ThD. In generation process 2, RD = 0 when the crossing time CT is 0, and RD = 1 when the crossing time CT is ThT or greater, and RD can be linearly changed in the range from 0 to 1 according to the value of the crossing time CT in the range from 0 to ThT. In generation process 3, RD = 0 when ABS(VS・VY) is greater than ThI, and RD = 1 when ABS(VS・VY) is 0, and RD can be linearly changed in the range from 0 to 1 according to the value of ABS(VS・VY) in the range from ThI to 0.

[0216] [7. Supplementary Notes] The embodiments disclosed herein are illustrative in all respects and are not restrictive. The scope of the present invention is indicated by the claims rather than by the embodiments described above, and includes the meaning of equivalents of the claims and all modifications within that scope.

[0217] 10 Radio wave sensor 14 Transmitting / receiving unit 15 Transmitting circuit 15a Transmitting antenna 16 Receiving circuit 16a Receiving antenna 17 Detection circuit 18 Housing 18a Transmitting / receiving surface 20 Pedestrian crossing 30, 300 Detection area 30_1, 300_1 Zebra area 30_2A, 30_2B, 300_2A, 300_2B Standby area 31O Point 31Y Projection center axis 31X Straight line 35 Radio wave irradiation area 50 Structure 51 Pole 52 Arm 60, 65 Roadway 61 Intersection 62a Inbound lane 62b Outbound lane 63a, 63b Sidewalk 71, 72, 73 Detection result 101 Processor 102 Non-volatile memory 103 Volatile memory 107 Communication Interface (Communication I / F) 110 Control Program 111 Setting Information 121 Identification Unit 122 Detection Unit 123 Tracking Unit 124 Generation Unit 125 Output Unit 131 First Identification Unit 132 Detection Unit 133 Tracking Unit 134 First Generation Unit 135 Acquisition Unit 136 Second Identification Unit 137 Second Generation Unit 138 Determination Unit 139 Estimation Unit 140 Output Unit 200 Setting Support Device 201 Processor 202 Non-volatile Memory 203 Volatile Memory 204 Input / Output Interface (I / O) 205 Graphics Controller 206 Communication Interface (Communication I / F) 210 Setting Support Program 211 Input Device 212 Display Device 221 Acquisition Unit 222 Identification Unit 223 Generation Unit 224 First Determination Unit 225 First estimation unit 226 First output unit 227 Second determination unit 228 Second estimation unit 229 Second output unit 230 First decision unit 231 Second decision unit 232 Third output unit CL1, CL2, CL3 Cluster t1, t2, t3, t4 Time point Pd1, Pd2, Pd3 Pedestrian V1, V2, V3, V4 Vehicle

Claims

1. A setting support device that assists in setting up a radio wave sensor for detecting objects in a detection area that includes at least a portion of a road by irradiating the area with radio waves, the setting support device comprising: an acquisition unit that acquires detection results of multiple objects by the radio wave sensor; and an identification unit that identifies low-reliability results, which are detection results with lower reliability than a standard, from the detection results of the multiple objects acquired by the acquisition unit.

2. The setting support device according to claim 1, further comprising: a generation unit that generates low reliability information relating to the occurrence of low reliability results based on the low reliability results identified by the identification unit.

3. The setting support device according to claim 2, wherein the low reliability information is the number of occurrences of the low reliability detection result per unit time.

4. The setting support device according to claim 2, wherein the low reliability information is the ratio of the number of detection results with low reliability to the total number of detection results for the object.

5. The setting support device according to any one of claims 2 to 4, further comprising: a first determination unit that determines whether the settings of the radio wave sensor are appropriate based on the low reliability information generated by the generation unit; and a first output unit that outputs the result of the determination by the first determination unit.

6. The setting support device according to claim 5, further comprising a first estimation unit that estimates the cause of the inappropriate setting of the radio wave sensor when the first determination unit determines that the setting of the radio wave sensor is inappropriate, and the first output unit outputs the cause estimated by the first estimation unit.

7. The setting support device according to claim 6, wherein the first estimation unit estimates the cause based on the location where the unreliable detection result occurred.

8. The setting support device according to any one of claims 2 to 7, further comprising: a second determination unit that determines whether or not the radio wave sensor is abnormal based on the low reliability information generated by the generation unit; and a second output unit that outputs the result of the determination by the second determination unit.

9. The setting support device according to claim 8, further comprising a second estimation unit that estimates the cause of the abnormality of the radio wave sensor when the second determination unit determines that the radio wave sensor is abnormal, and the second output unit outputs the cause of the abnormality estimated by the second estimation unit.

10. The setting support device according to claim 9, wherein the second estimation unit estimates the cause of the abnormality based on the location where the unreliable detection result occurred.

11. The setting support device according to any one of claims 8 to 10, wherein the second determination unit determines whether the radio wave sensor is abnormal based on the difference between first low reliability information generated by the generation unit based on the object detection result of the radio wave sensor at a first time point and second low reliability information generated by the generation unit based on the object detection result of the radio wave sensor at a second time point after the first time point.

12. The setting support device according to any one of claims 1 to 11, further comprising: a first determination unit that determines candidate areas which are candidates for exclusion areas to be excluded from the detection target of the radio wave sensor based on the detection position of the object in the low reliability result identified by the identification unit; and a third output unit that outputs the candidate areas determined by the first determination unit.

13. The setting support device according to claim 12, further comprising a second determination unit that determines the specified candidate area as the excluded area when the candidate area output by the third output unit is specified by the user.

14. An information processing method for processing information about a radio wave sensor that detects objects in a detection area by irradiating the detection area, which includes at least a portion of a road, the method comprising: acquiring detection results of a plurality of objects by the radio wave sensor; and identifying low-reliability results from the acquired detection results of the plurality of objects, which are detection results with reliability lower than a standard.

15. A computer program for processing information relating to a radio wave sensor that detects objects in a detection area by irradiating the detection area, which includes at least a portion of a road, the computer program for causing the computer to perform the steps of: acquiring detection results of a plurality of objects by the radio wave sensor; and identifying low-reliability results from the acquired detection results of the plurality of objects, which are detection results whose reliability is lower than a standard.

16. A radio wave sensor that detects objects in a detection area by irradiating the detection area, which includes at least a portion of a road, the radio wave sensor comprising: a storage unit that stores the detection results of a plurality of objects by the radio wave sensor; and a specification unit that identifies low-reliability results, which are detection results with lower reliability than a standard, from the detection results of the plurality of objects stored in the storage unit.

Images