DETECTION APPARATUS, DETECTION METHOD, AND PROGRAM

Abstract

To provide a detection device, a detection method, and a program that can accurately detect concave obstacles even when lower-side ground of the concave obstacles cannot be detected.SOLUTION: The above problem can be solved by a detection device (1) or the like including: a sensor device (10) that periodically detects heights of ground at a plurality of points around a mobile body (2); a detection range determination unit (24) that obtains a detection range (31) to the farthest point (35) located farthest from the sensor device (10) among the detected points; and a first determination unit (26) that determines that there is a concave obstacle (52) on the ground if there is a discontinuous detection area within the detection range (31).SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a detection device, a detection method, and a program, and more particularly to a detection device, a detection method, and a program capable of detecting a concave obstacle in the ground around a moving object. [Background technology]

[0002] For vehicles such as agricultural vehicles and construction vehicles that travel on relatively unimproved road surfaces such as footpaths, fields, and embankments, it is essential to detect concave obstacles such as steps and grooves on the road. One method for detecting such concave obstacles is to monitor the surrounding ground using a sensor device such as a radar or sonar attached to the vehicle. For example, Patent Document 1 proposes a technology that measures the altitude of the road surface in front of the vehicle using a laser radar device attached to the front of the vehicle, and sets candidate points for the concave obstacle based on the difference in altitude between a pair of measurement points corresponding to the ground above and below the concave obstacle and the gradient with respect to the horizontal. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2018-155694 A Summary of the Invention [Problem to be solved by the invention]

[0004] The above-mentioned method is based on the premise that two detection points with different heights can be detected. However, farm roads and construction sites are expected to have spaces with different height differences over a wide range in the distance direction, such as cliffs, and concave obstacles such as narrow grooves. On such road surfaces, the ground below the concave obstacle may be in the blind spot of the radar device, so it is not always possible to detect two detection points with different heights.

[0005] An object of the present invention is to provide a detection device, a detection method, and a program that can detect a concave obstacle with high accuracy even when a detection point on the ground below the concave obstacle cannot be obtained. [Means for solving the problem]

[0006] The above problem can be solved by a detection device that includes a sensor device that periodically detects the ground height at multiple points around a moving body, a detection range determination unit that determines the detection range to the farthest point from the sensor device among the detected points, and a first determination unit that determines that a concave obstacle exists on the ground if a discontinuous detection area is present within the detection range.

[0007] Here, the "moving object" does not necessarily mean a single vehicle, but includes a collection of moving objects such as a combination of a tractor and a trailer. The "sensor device" is a device that can be attached to a moving object, such as a radar device, LiDAR, or camera, to detect the position of the ground. The "discontinuous detection area" is an area in which detection points are not detected continuously over a predetermined range or more. Here, "continuous" means, for example, that two detection points adjacent in the distance direction are present within a predetermined distance range. Even if they are not connected in a plane (xy plane) approximately parallel to the ground, it is sufficient that the two detection points are within a predetermined distance in the radial direction centered on the sensor device. The "concave obstacle" means a terrain that includes a height difference that discontinuously decreases in height, such as a downward step, a groove, or a cliff.

[0008] It is desirable that the first judgment unit is configured to determine that when there is an area between the sensor device and the farthest point where the number of detected points is smaller than a first threshold value, this small area is a discontinuous detection area and a concave obstacle is present.

[0009] In addition, it is desirable that the first determination unit is configured to divide the area from the sensor device to the farthest point into a plurality of areas, determine the number of detected points for each area, detect the distance from the sensor device to the area where the number of detected points is smaller than a second threshold value and the length of the area, and determine that a concave obstacle is present if a discontinuous detection area is present based on the ratio of the distance to the area where the number of detected points is smaller than the second threshold value and the size of the area.

[0010] It is desirable that the detection device further include a second determination unit that determines the presence or absence of a concave obstacle on the ground based on temporal fluctuations in the detection range when the first determination unit does not determine the presence of a concave obstacle on the ground.

[0011] The second determination unit is preferably configured to determine that a concave obstacle is present when the determined detection range is smaller than the detection range within the most recent predetermined period.

[0012] In addition, it is desirable for the second judgment unit to determine that a concave obstacle is present when the ratio of the determined detection range to the average value of the detection range within a most recent specified period is smaller than a third threshold determined based on the speed of the moving body.

[0013] In addition, it is desirable that the sensor device detects the height of the ground by detecting reflected waves reflected by the ground, and the second determination unit is configured to compare the reception strength of the reflected waves within the obtained detection range with the reception strength within a specified most recent period to determine the presence or absence of a concave obstacle.

[0014] Moreover, it is preferable that the detection device further includes a third determination unit that determines the presence or absence of a concave obstacle based on a deviation in the continuous detection distance depending on the detection direction of the sensor device.

[0015] The above problem can also be solved by a detection method including the steps of periodically detecting the ground height at multiple points around a moving body using a sensor device, determining a detection range to the farthest point among the detected points that is located farthest from the sensor device, and determining that a concave obstacle is present on the ground if a discontinuous detection area is present within the detection range.

[0016] Furthermore, the above problem can also be solved by a program that enables a computer to implement the following functions: determining a detection range to the farthest point from the sensor device among multiple points around a moving body where the ground height is periodically detected by a sensor device; and determining that a concave obstacle exists on the ground if a discontinuous detection area is present within the detection range. Effect of the Invention

[0017] According to the present invention, even when the ground surface below the concave obstacle cannot be detected, the concave obstacle can be detected with high accuracy. [Brief description of the drawings]

[0018] [Figure 1] 1 is a schematic diagram of a detection device according to the present invention; [Diagram 2] FIG. 2 is an explanatory diagram of detection by a radar device. [Diagram 3] 2 is a flowchart of a detection method and program according to the present invention. [Figure 4] FIG. 13 is an explanatory diagram of a concave obstacle to be detected. [Diagram 5] 4 is a flowchart of detection by a radar device. [Figure 6] FIG. 13 is an explanatory diagram of height correction based on IMU information. [Figure 7] FIG. 13 is an explanatory diagram of height correction based on IMU information. [Figure 8] 1 is a flowchart of basic concave obstacle detection. [Figure 9] FIG. 1 is an explanatory diagram of concave obstacle detection 1. [Figure 10]1 is a flowchart of concave obstacle detection 1. [Figure 11] FIG. 1 is an explanatory diagram of a concave obstacle detection 1′ which is a modified example of the concave obstacle detection 1. [Figure 12] 1 is a flowchart of concave obstacle detection 1'. [Figure 13] FIG. 13 is an explanatory diagram of concave obstacle detection 2. [Figure 14] 13 is a flowchart of concave obstacle detection 2. [Figure 15] FIG. 13 is an explanatory diagram of concave obstacle detection 2. [Figure 16] FIG. 13 is an explanatory diagram of concave obstacle detection 2. [Figure 17] FIG. 13 is a schematic diagram showing a modified example of the detection device according to the present invention. [Figure 18] 11 is a flowchart of a modified example of the detection method and program according to the present invention. [Figure 19] FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] FIG. 1 shows a schematic configuration diagram of a detection device 1 according to an embodiment of the present invention. The detection device 1 includes a radar device 10, which is a type of sensor device, a detection range determination unit 24 that is connected to the radar device 10 and detects the detection range of the radar device 10, and three types of determination units (basic determination unit 25, first determination unit 26, and second determination unit 27) that detect concave obstacles. The detection device 1 is connected to a higher-level device 29 that receives the detection result of the concave obstacle and performs processes such as issuing an alarm and automatic steering to avoid the concave obstacle. Although FIG. 1 shows only one radar device 10, there may be multiple radar devices. Furthermore, the connection between the devices and the determination units may be configured by wire or wireless network.

[0020] The radar device 10 transmits and receives signals in a predetermined frequency band periodically (for example, every 50 ms), and is mounted on the upper part, lower part, or both of a vehicle 2 such as an agricultural machine, a construction machine, or a general automobile, to detect the condition of the ground around the vehicle. Specifically, as shown in FIG. 2, the position of the ground as seen from the vehicle 2 can be detected. FIG. 2(a) is a plan view of the vehicle 2 mounted with the sensor device 10, and FIG. 2(b) is a side view of the vehicle 2. The sensor device 10 detects a plurality of detection points 32, and detects three-dimensional position information including height for each detection point 32. Based on this group of detection points 32, the detection range 31 of the sensor device 10 can be determined. The detected information is used to detect steps such as ditches and cliffs near farm roads, and to prevent vehicles such as agricultural machines from falling off. It can also be used to detect the width of a farm road by detecting ditches on both sides.

[0021] The radar device 10 includes a signal generating unit 14 that generates a signal pattern of a periodic transmission signal, a transmission control unit 13 that is connected to the signal generating unit 14 and controls the operation of the signal generating unit 14, an oscillator 15 that is connected to the signal generating unit 14 and oscillates a transmission signal based on the signal pattern from the signal generating unit 14, a transmitting antenna 11 that is connected to the oscillator 15 and radiates a transmission wave oscillated by the oscillator 15 in a direction toward the ground, a receiving antenna 12 that receives a reflected wave of the transmission wave reflected by a target 40 on the ground, an ADC 17 that is connected to the receiving antenna 12 and converts a signal received by the receiving antenna 12 into a digital signal, a distance calculation unit 18 that calculates a relative distance of the target 40 with respect to the sensor device 10 based on the digital signal, a speed calculation unit 19 that calculates the speed of the vehicle 2 based on the digital signal, a threshold setting unit 20 that sets a detection threshold used by the detection unit 21, a detection unit 21 that detects a peak in the distance direction, an angle calculation unit 22 that calculates an arrival angle (horizontal angle, elevation angle) from the distance detected by the detection unit 21, and a correction unit 23 that corrects a deviation in the detection direction by using IMU information. The oscillator 15 and the ADC 17 can be directly connected to each other via a switch 16. The radar device 10 of this embodiment employs an FCM (Fast Chirp Modulation) method, but is not limited to the FCM method and may employ a pulse method.

[0022] The functions of the detection range determination unit 24, the basic determination unit 25, the first determination unit 26, and the second determination unit 27 are implemented by describing the functions of each unit in a program and executing the program by a computer, but each unit may be configured by individual hardware or software. The program is stored in the storage device of the computer.

[0023] Next, the operation of the detection device 1, i.e., the detection method and program according to the embodiment of the present invention, will be described with reference to Fig. 3 and Fig. 4. Fig. 3 is an overall flowchart, and Fig. 4 is an explanatory diagram for explaining the type of concave obstacle 52 determined by the determination unit.

[0024] First, the radar device 10 periodically detects the ground at a plurality of points around the vehicle 2, and provides the detection information to the detection range determination unit 24 (step 101). The detection information includes the speed of the vehicle 2, and three-dimensional position information including the height of the ground for each detected point (detection point), and the reception intensity of the reflected wave. The detection information is stored in the memory of the detection device 1 in chronological order for each frame corresponding to each period.

[0025] Next, the basic determination unit 25 executes a basic concave obstacle determination process (step 102). The basic concave obstacle determination process is carried out by calculating the average height of the detected ground surface h n is smaller than a predetermined threshold value, it is determined that a concave obstacle 52 is present. When a concave obstacle 52 is determined by the basic concave obstacle determination process (step 103), the fact that a concave obstacle 52 has been determined is transmitted to the higher-level device 29 (step 108), and the process ends.

[0026] If the concave obstacle 52 is not determined by the basic concave obstacle determination process, the detection range determination unit 24 obtains the detection range 31 from the radar device 10 to the farthest point located farthest from the radar device 10, and the first determination unit 26 executes the concave obstacle determination 1 process (step 104). The concave obstacle determination 1 is a process for determining a groove-shaped concave obstacle 52 within the detection range 31, as shown in FIG. 4(a). Since the groove-shaped concave obstacle 52 is a blind spot of the radar device 10, the detection point 32 is not obtained in the area of ​​the concave obstacle 52. The concave obstacle determination 1 determines that the concave obstacle 52 exists on the ground if there is a discontinuous detection area within the detection area between the radar device 10 and the farthest point. If the concave obstacle 52 is determined (step 105), the fact that the concave obstacle 52 has been determined is transmitted to the higher-level device 29 (step 108), and the process ends.

[0027] If the concave obstacle 52 is not determined by the process of concave obstacle determination 1, the second determination unit 27 executes the process of concave obstacle determination 2 (step 106). The concave obstacle determination 2 is a process for determining a cliff-like concave obstacle 52 within the detection range 31, as shown in FIG. 4(b). The cliff-like concave obstacle 52 is in the blind spot of the radar device 10, so there is no detection point 32 in an area farther than the cliff. If a concave obstacle is determined by the process of concave obstacle determination 2 (step 107), a notification that a concave obstacle has been determined is transmitted to the host device 29 (step 108), and the process ends. The concave obstacle determination 2 determines the presence or absence of a concave obstacle 52 on the ground based on the temporal fluctuation of the detection range of the radar device 10. If a concave obstacle 52 is determined (step 107), a notification that a concave obstacle 52 has been determined is transmitted to the host device 29 (step 108), and the process ends. On the other hand, if the concave obstacle 52 is not detected, a message indicating that a concave obstacle has not been detected is transmitted to the host device 29 (step 109), and the process ends.

[0028] Next, each process of radar detection (step 101), basic concave obstacle determination (step 102), concave obstacle determination 1 (104), and concave obstacle determination 2 (106) will be described in detail.

[0029] 5 is a flowchart showing the operation of radar detection (step 101). First, the transmission control unit 13, the signal generation unit 14, and the oscillation unit 15 generate a transmission signal of a predetermined frequency, and perform a transmission process in which a transmission wave is radiated from the transmission antenna 11 toward the ground in front of the vehicle 2 (step 701). Next, the transmission wave is reflected by one or more targets (ground) 40, and the reflected wave is received as a reception signal by the reception antenna 12, and a reception process is performed in which the ADC 17 converts the reflected wave into a digital signal (step 702).

[0030] Next, the distance calculation unit 18 performs distance calculation to obtain the relative distance between the radar device 10 and the target (ground) 40 based on the digital signal (step 703). Also, the speed calculation unit 19 performs speed calculation based on the digital signal (step 704). Specifically, the speed calculation unit 19 calculates relative speed information with respect to one or more objects based on the received signal. The speed of the object can be detected by the speed calculation. After that, the threshold setting unit 20 sets a detection threshold for detecting the peak used by the detection unit 21 (step 705). Specifically, the threshold is adjusted based on the mounting position, height, angle, etc. of the receiving antenna 12. Then, the detection unit 21 detects the peak in the distance direction using a predetermined method such as CFAR (Constant False Alarm Rate) (step 706). Next, the angle calculation unit 22 performs angle calculation processing to obtain the arrival angle (horizontal angle, elevation angle) of the reflected wave at the distance detected by the detection unit 21, and calculates the direction in which the target (ground) 40 is located in a virtual space with the position of the radar device 10 as the origin and the radiation direction as the reference axis.

[0031] Furthermore, the correction unit 23 corrects the deviation of the detection position, which occurs when the radar device is tilted due to the influence of unevenness of the ground, by using IMU (Interial Measurement Unit) information. As shown in A of FIG. 6, when the vehicle 2 is traveling on a flat ground, the positional relationship between the radar device 10 and the ground 3 does not change, so the height can be detected without any deviation. As is clear from the graph in the figure, the detection point 32 is located within a predetermined range 60. In contrast, when the vehicle 2 is traveling on an uneven ground as shown in B of FIG. 6, the position of the target (ground) 40 obtained by the radar device 10 is determined by the relative position from the radar, so a deviation occurs. For example, when the radar device 10 faces upward, the ground is detected as lower than the actual position, and the detection point 33 is detected as being significantly shifted from the range 60 where it should be located. In order to eliminate this deviation, the change in the vehicle's attitude (amount of three-dimensional movement) is recognized from the vehicle's IMU information, and the coordinate information seen from the radar device 10 is converted to global coordinates (coordinates with the start position as the origin). 7 shows (a) the detection result before IMU correction by the correction unit 23 and (b) the detection result after the correction. It can be seen that the detection point 33, which was significantly deviated from the range 60 before the correction, has been corrected to the detection point 34 located within the range 60 by the correction.

[0032] Through the above processing, the radar device 10 can detect detection information including three-dimensional position information of each detection point, the reception intensity of the reflected wave, and the speed of the target. The detection information of the radar device 10 is stored in the memory of the detection device 1 in chronological order for each frame corresponding to each period.

[0033] Next, detailed processing of the basic concave obstacle determination (step 102) will be described with reference to Fig. 8. The basic concave obstacle determination is performed by a computer (processor) constituting the basic determination unit 25 executing a program stored in a memory.

[0034] First, the basic determination unit 25 reads out the detection information of the latest frame (hereinafter, referred to as the "current frame") from the detection information stored in the memory of the detection device 1, and obtains the height information included in the three-dimensional position information for each detection point 32. Then, the average value h n Next, the basic determination unit 25 calculates a threshold value th h (Step 202). For example, the threshold th h As shown in equation (1), the actual ground height g is calculated based on the radar installation height, etc. h The value is set within a specified range (t) based on the standard. th h =f1(g h <f1<g h +t) ···(1)

[0035] Next, the basic determination unit 25 calculates the average height of the target detected in the current frame, h n and the threshold th h The average height of the target detected in the current frame is compared with h n is the threshold th h If the average height of the target detected in the current frame is lower than h n is the threshold th h If so, it is determined that there is no concave obstacle 52 on the ground 3 within the detection range 31, and the process ends.

[0036] Next, the detailed process of concave obstacle determination 1 (step 104) will be described with reference to FIG. 9 and FIG. 10. Concave obstacle determination 1 is performed by a computer (processor) constituting the first determination unit 26 executing a program stored in a memory. Concave obstacle determination 1 is a process for determining a groove-like concave obstacle 52 in the detection range 31. Specifically, as shown in FIG. 9, the detection range 31 is divided into areas 61, 62, and 63 for each predetermined search range r in the distance direction from the radar device 10, and the number of detection points 32 in the areas is counted in order from the area closest to the radar device 10. When there is a discontinuous detection area smaller than a predetermined threshold, this small area is determined to be a concave obstacle. The detection range 31 is defined by the distance from the radar device 10 to the farthest point 35 farthest from the radar device 10.

[0037] Next, a specific processing procedure will be described based on the flowchart of concave obstacle judgment 1 shown in FIG. 10. First, the first judgment unit 26 initializes a distance index d to 0 (step 301) and sets a predetermined search range r (step 302). Next, the first judgment unit 26 calculates the distance w to the farthest point. max (Step 303). The distance to the farthest point w max As shown in FIG. 9, is the distance to the detection point 35 that is located farthest from the radar device 10 among the detection points 32.

[0038] Next, the first determination unit 26 determines the number of detection points cnt that exist in the region from the distance index d to d+r. d At this point, since the distance index d=0, the number of detection points cnt that exist in the area 61 from the radar device 10 to the search range r is counted (step 304). d Then, the first determination unit 26 counts the number of detection points cnt d and a given threshold th cnt The number of counted detection points cnt is compared (step 305). d is a given threshold th cntIf the number of detected points counted is less than cnt, it is determined that a discontinuous detection area exists in the area 61 and that a concave obstacle 52 exists (step 308), and the process ends. d is a given threshold th cnt In the above cases, it is determined that no concave obstacle 52 exists in the region 61. Next, the first determination unit 26 adds the distance index d to the search range r (step 306), and calculates the distance index d after the addition and the distance to the farthest point w max (Step 307). The distance index d after the addition is compared with the distance w max If so, return to step 304.

[0039] At this point, since the distance index d=r, the first determination unit 26 determines the number of detection points cnt that exist in the region 62 whose distance from the radar device 10 is from r to 2r. d (Step 304). Then, the first determination unit 26 counts the number of detection points cnt d and a given threshold th cnt The number of counted detection points cnt is compared (step 305). d is a given threshold th cnt If the number of detected points counted is less than cnt, it is determined that a discontinuous detection area exists in the area 62 and that a concave obstacle 52 exists in the area 62 (step 308), and the process ends. d is a given threshold th cnt In the above cases, it is determined that no concave obstacle 52 exists in the region 62. Next, the first determination unit 26 adds the distance index d by the search range r (step 306), and calculates the distance index d after the addition and the distance w to the farthest point. max (Step 307). The distance index d after the addition is compared with the distance w max If so, return to step 304.

[0040] In this way, the concave obstacle 52 is determined, or the distance index d after the addition is the distance w maxIf the distance index d after the addition is greater than the distance w to the farthest point, the process of steps 304 to 307 is repeated. max If it is determined that it is greater than the threshold value, it is determined that there is no discontinuous detection area within the detection range 31, and therefore that there is no concave obstacle 52 on the ground 3 within the detection range 31, and the process ends.

[0041] Next, the process of concave obstacle determination 1', which is a modified example of concave obstacle determination 1, will be described with reference to Figs. 11 and 12. Concave obstacle determination 1' is performed by the computer (processor) constituting the first determination unit 26 executing a program stored in memory. Concave obstacle determination 1' is a process for determining a groove-like concave obstacle 52 within the detection range 31, and is similar to concave obstacle determination 1 in that it divides the area from the radar device 10 to the farthest point 35 into a plurality of regions and determines the number of detected points for each region. However, as shown in Fig. 11, a histogram showing the number of detected points with respect to distance is created, and the number of detected points is calculated from the radar device 10 when the number of detections exceeds a predetermined threshold th cnt’ Distance w to the smaller region 66 c , and the length w of the region 66 d Detect the distance w c and length w d The difference is that the presence or absence of the concave obstacle 52, which is a discontinuous detection area, is determined based on the ratio of the distance w c and its length w d By making the determination based on the ratio of the above, it is possible to prevent a narrow dead zone from being erroneously determined to be a concave obstacle 52, and it is possible to more accurately determine whether or not there is a concave obstacle 52.

[0042] Next, a specific process will be described based on the flowchart of concave obstacle determination 1' shown in Fig. 12. First, the first determination unit 26 divides the area from the radar device 10 to the farthest point 35 into a plurality of areas 61, 62, and 63, obtains the number of detection points for each area, and calculates the number of detection points cnt with respect to the distance r as shown in Fig. 11. d Next, the first determination unit 26 creates a histogram showing the number of detected points counted cnt for each region. dand a given threshold th cnt’ The number of detection points counted by comparing with d and a given threshold th cnt’ It is determined whether there is a region 66 that is smaller than the detected point (step 402). d is a given threshold th cnt’ If the smaller area 66 is not detected, it is determined that there is no discontinuous detection area within the detection range 31, and that there is no concave obstacle 52 on the ground 3 within the detection range 31, and the process ends.

[0043] On the other hand, the number of detection points cnt d is a given threshold th cnt’ When a region 66 that is smaller than the position 10 of the radar device is detected, the first determination unit 26 determines the distance w c (step 403), and the length w d (Step 404). Then, the first determination unit 26 determines whether the area 66 can be evaluated as a discontinuous detection area. Specifically, the distance w from the position 10 of the radar device to the area 66 is c and the length of region 66 w d The ratio of w d / w c and a given threshold th hist and (step 405). d Since the value varies depending on the radar installation height and angle, the threshold value th hist The ratio w may be adjusted depending on the position and angle of the radar device. d / w c is the threshold th hist If the ratio w is greater than 1, it means that the area 66 is large, so it is determined that the area 66 can be evaluated as a discontinuous detection area, and it is determined that a concave obstacle 52 is present in the detection range 31 (step 406), and the process ends. d / w c is the threshold th hist In the following cases, the area 66 cannot be evaluated as a discontinuous detection area, it is continuously detectable across the detection range 31, there is no discontinuous detection area within the detection range 31, and therefore it is determined that there is no concave obstacle 52 on the ground 3 within the detection range 31, and the process ends.

[0044] Next, the detailed process of the concave obstacle determination 2 (step 106) will be described with reference to FIG. 13 and FIG. 14. The concave obstacle determination 2 is performed by the computer (processor) constituting the second determination unit 27 executing a program stored in the memory. The concave obstacle determination 2 is a process for determining the presence or absence of a cliff-like concave obstacle 52 in the detection range 31 based on the temporal variation of the detection range when the first determination unit 26 does not determine the presence of a concave obstacle 52 on the ground. That is, when the detection range obtained from the detection point of the current frame is smaller than the detection ranges of a plurality of frames within a predetermined period of time, it is determined that the concave obstacle 52 exists. More specifically, when the ratio of the detection range obtained from the detection point of the current frame to the average value of the detection ranges of a plurality of frames within a predetermined period of time is smaller than a third threshold value determined based on the speed of the moving object, it is determined that the concave obstacle 52 exists. 13(a), when the detection range 31 does not include a cliff-like concave obstacle 52, the farthest point 35 is located near the boundary of the detection range 31. However, when the detection range 31 includes a cliff-like concave obstacle 52, as shown in FIG. 13(b), since no detection point can be obtained in an area farther away than the cliff-like concave obstacle 52, the farthest point 35 is located closer to the actual detection range 39, and the detection range 31 becomes narrower. For this reason, the concave obstacle 52 can be detected by detecting the time-series changes in the detection range 31.

[0045] Next, a specific process procedure of concave obstacle judgment 2 will be described with reference to the flowchart in Fig. 14. First, the second judgment unit 27 obtains the detection range within a predetermined most recent period (step 501). That is, as shown in Fig. 15(a), going back n-1 frames from the current frame (frame No. n), the second judgment unit 27 obtains the detection range w1 to w2 from the detection points successively detected by the sensor device 10 for the detection points of each of the past frames (frame Nos. 1 to n-1). n-1 Next, the second determination unit 27 calculates the detection range w1 to w n-1 The average value of w ave(step 502), and the detection range w n and the average detection range of the past frames w ave The ratio of n / w ave In the above-mentioned steps 501 to 503, all the detection ranges w1 to w n-1 Using the average value w ave However, only a part of the frames is used to find the average detection range w ave You may ask for:

[0046] Next, the second determination unit 27 determines whether or not there is a change in the detection range by using a change determination threshold th f (Step 504). In the case where there is a ditch or a cliff on the ground, the detection range w n Since the ratio w becomes smaller, n / w ave is 0 <w n / w ave <1. The closer to 0, the closer the object is to a ditch or cliff. Therefore, the fluctuation judgment threshold th f is a predetermined value smaller than 1, as shown in equation (2). th f =f2(0 <f2<1) ···(2) Here, th f is not limited to a fixed value, and may be multiplied by a coefficient k according to the speed of the vehicle 2 on which the radar device 10 is mounted, as in equation (3). th f =f2×k (0 <f2×k<1) ··· (3) That is, the second determination unit 27 reads out a coefficient k from a table set for each vehicle speed as shown in Fig. 16 based on the vehicle speed v of the vehicle 2 included in the detection information of the radar device 10. The coefficient k is larger as the vehicle speed is faster and is smaller as the vehicle speed is slower. Next, the second determination unit 27 performs a calculation as shown in formula (3) to determine the fluctuation determination threshold th based on the coefficient k. f The coefficient k may be determined by calculating using a function of the vehicle speed v without referring to a table.

[0047] Next, the second determination unit 27 detects the detection range wn and the average detection range of the past frames w ave The ratio of n / w ave and the fluctuation judgment threshold th f The ratio w is compared (step 505). n / w ave is the fluctuation judgment threshold th f If the above is true, it is determined that there is no time-series variation in the detection range 31 and that there is no concave obstacle 52 on the ground 3 within the detection range 31, and the process ends.

[0048] On the other hand, n / w ave is the fluctuation judgment threshold th f If the difference is smaller than the previous frame, a time-series variation in the detection range 31 is detected. Here, the time-series variation in the detection range 31 may occur not only due to the presence of the concave obstacle 52, but also due to a change in the intensity of the reflected wave from the ground 3 caused by other factors such as the material of the ground 3. For this reason, the time-series variation in the reception strength of the reflected wave included in the detection information of the radar device 10, i.e., the reception strength of the reflected wave at the detection point in the current frame, is compared with the reception strength of the reflected wave at the detection point in the most recent predetermined period. More specifically, the ratio between the average value of the reception strength of the reflected wave at the detection point in the current frame and the average value of the reception strength at the detection point in the most recent predetermined period is compared with a predetermined threshold. If the change in the reception strength is smaller than that in the previous frame, it is determined that the influence of the reception strength is small, and it is possible to determine that the time-series variation in the detection range 31 is due to the concave obstacle 52, and the concave obstacle 52 can be detected.

[0049] In order to determine the influence of the reception strength described above, the second determination unit 27 first calculates the average reception strength amp of the reflected wave at the detection point in the current frame. n Next, the reception strength during the most recent predetermined period is calculated, and the average reception strength of the past frames, amp aveThat is, as shown in FIG. 15B, by going back n-1 frames from the current frame (frame No. n), the average values ​​amp1 to amp2 of the reception strength of the reflected wave at the detection point in each frame are calculated. n-1 Next, the second determination unit 27 calculates the reception intensities amp1 to amp n-1 Average value of amp ave In this case, the average value of all the reception strengths for the most recent n-1 frames, amp1 to amp n-1 Using the average value amp ave Alternatively, the average value amp can be calculated using only a portion of the frames. ave You may ask for:

[0050] Next, a fluctuation judgment threshold th for judging whether or not the reception strength fluctuates is set. p (Step 508). Then, the second decision unit 27 sets the average value amp n and the average reception strength of past frames amp ave Comparison with n / amp ave The fluctuation judgment threshold th p (Step 509). n / amp ave is the fluctuation judgment threshold th p If the ratio is more than this, there is no time-series variation in the reception strength, and the variation in the detection range 31 detected in step 505 is due to the concave obstacle 52 on the ground 3. It is determined that the concave obstacle 52 is present on the ground 3 within the detection range 31 (step 510), and the process ends. n / amp ave is the fluctuation judgment threshold th p If it is smaller, it is determined that the time series fluctuation of the detection range 31 detected in step 505 is likely to have been caused by other factors such as the material of the ground 3, and it is determined that there is no concave obstacle 52 on the ground 3 within the detection range 31, and the process ends.

[0051] As described above, the detection device, detection method, and program according to the present invention have been described with reference to the configuration and operation of the detection device 1 according to an embodiment of the present invention. If there is a discontinuous detection area within the detection range, it is possible to determine that a concave obstacle exists on the ground, and thus to accurately detect the concave obstacle even if the ground below the concave obstacle cannot be detected.

[0052] Next, modified examples of the detection device, detection method, and program according to the present invention will be described with reference to Figs. 17 to 19. Fig. 17 is a schematic diagram of the detection device 4, and Fig. 18 is a flowchart showing the operation of the detection device 4. The detection device 4 differs from the configuration of the detection device 1 described above (Fig. 1) in that it includes a third determination unit 28 connected to the radar device 10 and a higher-level device 29, but the other configuration is the same as that of the detection device 1. The operation of the detection device 4 differs from the operation of the detection device 1 described above (Fig. 3) in that steps 110 and 111 in Fig. 18 are provided, but the other operations are the same as those of the detection device 1. The third determination unit 28 is configured to implement steps 110 and 111 of the flowchart shown in Fig. 18.

[0053] In the detection device 1, all detection points 32 are used regardless of the detection direction of the radar device 10 to determine the presence or absence of a concave obstacle 52 within the detection range 31. In contrast, as shown in FIG. 19(a), the detection device 4 divides the detection range 31 into a plurality of regions 71, 72 for each predetermined angle range θ centered on the radar device 10, and calculates the continuous region distance w c The presence or absence of a concave obstacle 52 is determined from the above. When traveling on a normal flat road, the continuous detection distance does not vary greatly depending on the detection direction, but if there is a groove on the side of the vehicle, for example, the continuous detection distance in the direction of the groove will be shorter while the continuous detection distance in the direction of no groove will remain unchanged, resulting in a bias in the continuous detection distance depending on the detection direction. Therefore, by detecting this bias, it is possible to determine whether or not there is a concave obstacle 52 nearby.

[0054] Next, the operation of the detection device 4 will be described with reference to Fig. 18. Note that the same operations as those of the detection device 1 will be omitted by using the same references. After the first determination unit 26 determines whether there is a concave obstacle 52 (steps 104, 105), the third determination unit 28 divides the detection range 31 into a plurality of regions 71, 72 at predetermined angles θ with the radar device 10 as the center, and calculates the continuous region distance w in each region based on the detection points in each region. c The continuous detection distance w is calculated (step 110). c is the distance w of the concave obstacle judgment 1' c In other words, the number of detection points cnt for each of the regions 71 and 72 divided in the angular direction and further divided in the radial direction can be calculated by the same method. d is a given threshold th cnt’ The distance from the area where the detection range becomes smaller to the position of the radar device 10 is the continuous detection range w c For each region 71 and 72, the distance w c By calculating the continuous detection distance w c A histogram of the values ​​can be created.

[0055] Next, the third determination unit 28 determines whether or not there is a bias in the continuous detection distance depending on the detection direction of the radar device 10 (step 111). The detection device 4 uses a statistical method such as calculating a variance value to determine the bias due to the detection direction, but the bias may be determined using other methods. For example, the change amount or rate of change in the distance detected for each adjacent angle may be calculated, and if it is determined that there is a change equal to or greater than a predetermined threshold, it may be determined that there is a bias. If there is a bias due to the detection direction, it is determined that a concave obstacle 52 exists, and the fact that a concave obstacle has been determined is transmitted to the upper device 29 (step 108), and the process ends. On the other hand, if there is no bias due to the detection direction, the second determination unit 27 executes concave obstacle determination 2 (steps 106, 107).

[0056] As described above, the detection device 4 detects a concave obstacle taking into consideration the detection direction, and can therefore accurately detect even a concave obstacle located in a specific direction relative to the vehicle, such as a side gutter.

[0057] The detection device, detection method, and program according to the present invention have been described above, but the present invention is not limited to the above-described embodiments, and includes all aspects included in the concept of the present invention and the scope of the claims. [Explanation of symbols]

[0058] 1, 4 Detection device 2. Mobile objects (vehicles) 3 ground 10 Sensor equipment (radar equipment) 11 Transmitting Antenna 12 Receiving antenna 13 Transmission control section 14 Signal Generation Unit 15 Oscillator 16 Switch 17 ADC 18 Distance calculation section 19 Speed ​​calculation section 20 Threshold setting unit 21 Detection unit 22 Angle calculation section 23 Correction section 24 Detection range determination unit 25 Basic judgment section 26 1st Judgment Section 27 Second Judgment Section 28 Third Judgment Section 29 Upper device 31, 39 Detection range 32, 33, 34, 36, 37 Detection points 35 farthest point 40 Target (ground) 51 Upper ground 52 Concave Obstacle 53 Lower ground 61, 62, 63, 66, 71, 72 area

Claims

1. A sensor device that periodically detects ground heights at a plurality of points around a moving object; a detection range determination unit that determines a detection range to a farthest point that is located farthest from the sensor device among the detected points; a first determination unit that determines that a concave obstacle is present on the ground if a discontinuous detection area is present within the detection range; Equipped with The first determination unit is Dividing an area from the sensor device to the farthest point into a plurality of areas, and determining the number of the points detected for each area; Detecting a distance from the sensor device to a region in which the number of the detected points is smaller than a first threshold value, and a length of the region; determining that the concave obstacle is present if the discontinuous detection area exists based on a ratio of a distance to the area where the number of the detected points is smaller than a first threshold value to a length of the area; Detection device.

2. The detection device according to claim 1 , further comprising a second determination unit that determines the presence or absence of a concave obstacle on the ground based on temporal fluctuations in the detection range when the first determination unit does not determine the presence of a concave obstacle on the ground.

3. The detection device according to claim 2 , wherein the second determination unit determines that the concave obstacle is present when the obtained detection range is smaller than the detection range within a most recent predetermined period.

4. 3. The detection device according to claim 2, wherein the second determination unit determines that the concave obstacle is present when a ratio of the determined detection range to an average value of the detection range within a most recent specified period is smaller than a second threshold determined based on a speed of the moving body.

5. The sensor device detects the height of the ground by detecting a reflected wave reflected by the ground, The second determination unit compares the reception strength of the reflected wave within the obtained detection range with the reception strength within a recent predetermined period to determine the presence or absence of the concave obstacle.

5. A detection device according to any one of claims 2 to 4.

6. The detection device according to claim 1 , further comprising a third determination unit that determines the presence or absence of the concave obstacle based on a deviation in continuous detection distance depending on a detection direction of the sensor device.

7. periodically detecting ground heights at a plurality of points around the moving object by a sensor device; determining a detection range from the detected point to a farthest point located farthest from the sensor device; determining that a concave obstacle is present on the ground if a discontinuous detection area is present within the detection range; Including, The determining step includes: Dividing an area from the sensor device to the farthest point into a plurality of areas and calculating the number of the points detected for each area; Detecting a distance from the sensor device to an area where the number of detected points is less than a first threshold, and a length of the area; determining that the concave obstacle is present if the discontinuous detection area exists based on a ratio of a distance to the area where the number of the detected points is smaller than a first threshold value to a length of the area; Including, Detection methods.

8. On the computer, A function of calculating a detection range from a sensor device to a farthest point located farthest from the sensor device among a plurality of points around the moving object, the ground height of which is periodically detected by the sensor device; a function of determining that a concave obstacle is present on the ground if a discontinuous detection area is present within the detection range; A program for achieving the above, The function of determining is A function of dividing an area from the sensor device to the farthest point into a plurality of areas and calculating the number of the points detected for each area; A function of detecting a distance from the sensor device to an area in which the number of the detected points is smaller than a first threshold value, and a length of the area; a function of determining that the concave obstacle is present if the discontinuous detection area exists based on a ratio of a distance to an area where the number of the detected points is smaller than a first threshold value to a length of the area; Including, program.

Images