Object detection system

Abstract

To provide an object detection system capable of detecting an object at high accuracy.SOLUTION: An object detection system includes: a cart 4 in which a plurality of reflectors 51-54 having mutually different reflection factors, as respective markers 5, are provided in prescribed arrays on a plurality of surfaces directed in different directions of exterior surfaces; a detection device (1) for detecting the reflectors 51-54; and a management control device (2) for detecting the cart 4 from an array state of the detected reflectors 51-54. The cart 4 comprises surfaces directed in at least three different directions, and the reflectors 51-54 having different reflection factors respectively are arranged on each surface.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to a detection system that detects an object, and more particularly to an object detection system that is equipped with a detection device such as a LiDAR and is capable of detecting objects with high accuracy. [Background technology]

[0002] In recent years, services have been provided in which hand carts for transporting luggage (hereinafter, abbreviated as carts) and wheelchairs are rented out within certain facilities such as shopping malls. In such services, people who use a cart may not return it to a designated location, which may result in people waiting for a cart to be available to use it, or the abandoned cart may become an obstacle to pedestrians and vehicles. Therefore, there is a need to detect unreturned carts left within the facility and quickly return them to the designated location. It is time-consuming and costly for a caretaker to walk around the facility looking for unreturned carts, and is therefore inefficient.

[0003] Patent Document 1 proposes a technology for detecting a moving object by detecting a marker provided on the moving object with a detection device such as LiDAR, thereby detecting the moving object and further detecting the size and moving direction of the moving object. For example, LiDAR projects infrared light while scanning, and detects an object based on the reflected light reflected by the object. In particular, Patent Document 1 detects the marker by setting the marker to a predetermined reflectance and comparing the light intensity of the reflected light of the marker detected by LiDAR with a predetermined threshold value. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] International Publication No. 2020 / 195607 Summary of the Invention [Problem to be solved by the invention]

[0005] Therefore, in order to detect unreturned objects in the above-mentioned facilities, it is possible to apply the detection technology of Patent Document 1. That is, a marker with a predetermined reflectance is provided on a part of the cart, and the cart is detected by detecting the marker using LiDAR. However, if an object with a reflectance similar to that of the marker is present near the cart to be detected, it is difficult to distinguish between the reflected light from the marker of the cart to be detected and the reflected light from other objects, making it difficult to detect the cart with high accuracy.

[0006] In particular, since the marker in Patent Document 1 is configured to have a predetermined reflectance, if another object has a portion with a similar reflectance, the other object may be mistakenly detected as a marker. In other words, if the technology in Patent Document 1 is applied to the detection of an unreturned cart in the above-mentioned facility, the cart may be mistakenly detected due to reflected light from other objects in the facility, such as bicycles, vehicles such as automobiles, and other structures, resulting in a decrease in detection accuracy. This also applies when the object is a wheelchair or other equipment.

[0007] Patent Document 1 also proposes a technology in which markers are formed into a one-dimensional or two-dimensional array pattern and detected by detecting this array pattern. In order to detect this type of marker with LiDAR, it is necessary to form the marker into an array pattern that is at least larger in size than the detection resolution of LiDAR. Therefore, when markers are arranged on a cart, if there are restrictions on the size of the markers, it becomes difficult to use markers with an array pattern that can be detected by LiDAR, and from this point of view, it becomes difficult to perform highly accurate detection.

[0008] An object of the present invention is to provide an object detection system capable of detecting an object with high accuracy. [Means for solving the problem]

[0009] The object detection system of the present invention includes an object having a plurality of reflective materials, each having a different reflectance, arranged in a predetermined arrangement on a plurality of surfaces facing different directions on the outer surface of the object, a detection device that detects the plurality of reflective materials with different reflectances, and a management control device that detects the object from the arrangement state of the detected plurality of reflective materials. Here, it is preferable that the plurality of reflective materials include at least three reflective materials each having a different reflectance. It is also preferable that the object has at least three surfaces facing different directions, and each surface is provided with a reflective material with a different reflectance. Furthermore, the reflective material may be configured to have a required reflectance by combining a plurality of reflective materials with different reflectances at a required area ratio.

[0010] In the present invention, the detection device is preferably configured as a LiDAR that projects light onto the reflector, receives the light reflected by the reflector, and outputs detection data including at least information on the position and reflectance of the reflector. Also, the management control device is preferably configured to detect each reflector based on the detection data detected by the detection device and a number of detection thresholds corresponding to the number of different reflectances. Effect of the Invention

[0011] According to the present invention, an object is detected by utilizing a plurality of reflective materials with different reflectivities, thereby making it possible to diversify the arrangement pattern of the detection data when an object is detected, thereby preventing confusion with objects present in the vicinity, and providing an object detection system that can detect objects with high accuracy. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a conceptual diagram of the overall configuration of a cart detection system. [Diagram 2] FIG. [Diagram 3] FIG. 2 is a conceptual diagram of a management control device. [Figure 4] A conceptual diagram of a recovery robot. [Diagram 5] 1A and 1B are external views of the cart, where (a) is a perspective view, (b) is a plan view, and (c) is an enlarged view of a portion of the reflective material with a different shape. [Figure 6] Detection flow diagram. [Figure 7] FIG. 11 is a schematic diagram for explaining a detection flow. [Figure 8] Schematic diagram showing the correlation between detection data and binary data. [Figure 9] FIG. [Figure 10] 11A-11C are plan views of markers in other configurations. [Figure 11] FIG. 13 is a side view of a cart with markers of another configuration. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Next, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a conceptual diagram of an embodiment in which the object detection system of the present invention is configured as a detection system for detecting carts used in facilities such as shopping malls. Users of the facility can use carts 4 placed in designated cart parking areas 100, and after use, they return the carts to the designated cart parking areas 100. However, some carts are left abandoned in the facility without being returned. In order to detect such abandoned carts, a marker 5 is provided on the cart 4 as described below, and a detection device 1 for detecting the marker 5 and detecting the cart 4 is provided in the facility. Note that there may be multiple designated parking areas 100 for the carts 4, and when there are multiple types of carts 4, the designated parking areas may differ for each type.

[0014] Regarding the detection device 1, lighting fixtures 110 are provided at multiple locations within the facility, and a detection device 1 is disposed for each of these lighting fixtures 110. The lighting fixtures 110 have illuminators 112 supported at the upper ends of support poles 111, and the detection device 1 is supported within the case of the illuminators 112 or on the upper part of the support pole 111 separately from the illuminators 112. Each detection device 1 is set so that a predetermined area within the facility is its detection area, and the multiple detection areas are combined to cover substantially the entire facility. In this case, it is preferable that the detection areas of adjacent detection devices 1 are set so as to partially overlap each other.

[0015] A control room 120 where a caretaker is stationed is also provided within the facility. This control room 120 is equipped with a detection control device 2 that detects carts 4 based on detection signals detected by multiple detection devices 1 and detects cart information of the detected carts 4. This detection control device 2 is capable of displaying the detected cart information on a monitor 23, and the caretaker can check the cart information displayed on this monitor 23 and collect the cart 4 in the designated storage area 100.

[0016] 2 is a conceptual diagram of the detection device 1, and includes a LiDAR 11 as a detection unit. An existing LiDAR 11 can be used, but here, the device includes a light projecting unit 12 that projects infrared light in pulses and scans the projected infrared light over a detection area, a light receiving unit 13 that receives light reflected by an object or the like from the projected light, and a signal processing unit 14 that photoelectrically converts the light received by the light receiving unit 13, executes required processing, and outputs a detection signal. The detection device 1 also includes a detection communication unit 15 that transmits the detection signal output from the LiDAR 11 to the outside. This detection communication unit 15 is capable of wireless or wired communication with the detection control device 2 provided in the management room 120 described above.

[0017] 3 is a conceptual diagram of the detection control device 2 provided in the management room 120. The detection control device 2 includes a management communication unit 21 that receives detection signals transmitted from a plurality of detection devices 1, an information detection unit 22 that processes the received detection signals using a predetermined algorithm to detect a cart 4 and further detects the cart information, and a monitor 23 that displays the detected cart information. Here, the cart information includes attributes of the detected cart 4, i.e., the cart position (distance, direction), the cart orientation, and the designated cart storage location. The management communication unit 21 is connected to the detection communication unit 15 wirelessly or by wire.

[0018] In this embodiment, a recovery robot 3 capable of automatic travel is provided in the management room 120 shown in FIG. 1. As shown in the conceptual configuration in FIG. 4, this recovery robot includes a travel drive unit 31, a control unit 32, and a robot communication unit 33. The robot communication unit 33 is wirelessly connected to the management communication unit 21 and is capable of receiving cart information output from the information detection unit 22 of the management control device 2. The control unit 32 controls the travel drive unit 31 based on the received cart information, and drives the wheels 34 to automatically travel toward the cart 4. In addition, an algorithm is set so that the robot automatically couples to the cart using the arm 35, travels to the designated storage area 100, and then detaches from the cart 4 at the designated storage area 100 and returns to the management room 120.

[0019] 5 is an external view of the cart 4 as an object to be detected, (a) being a perspective view and (b) being a plan view. Casters 42 are attached to the lower part of a frame 41, and a basket section 43 for storing luggage is attached to a position near the upper part of the frame 41. The upper part of the frame 41 is configured as a handle 44 for a person to push the cart. The basket section 43 is formed in a rectangular box shape with an opening at the top, and usually each of its four sides, i.e., the front, rear, left and right sides, is formed in a mesh or lattice shape. A board-shaped (plate-shaped) reflective material is attached to each of the four sides of the basket section 43, and these reflective materials constitute the marker 5.

[0020] The reflecting material constituting the marker 5 is composed of a front reflecting material 51, a rear reflecting material 52, a right reflecting material 53, and a left reflecting material 54 attached to each of the four side surfaces of the basket portion 43. The left and right direction is the direction when looking forward from the handle 44 side of the cart 4. Each reflecting material 51 to 54 is attached over almost the entire surface of the corresponding side surface, but the front reflecting material 51 and the rear reflecting material 52 are attached in a state where both left and right ends of the reflecting material 51 and the rear reflecting material 52 protrude to the left and right sides by a required dimension, respectively. In other words, the right reflecting material 53 and the left reflecting material 54 are attached in a state where each part of the front reflecting material 51 and the rear reflecting material 52 is advanced to the left and right ends, respectively. Therefore, when the marker 5 is viewed from the left side or right side of the cart 4, each part of the front reflecting material 51 and the rear reflecting material 52 is visually recognized together with the left and right parts of the right reflecting material 53 and the left reflecting material 54.

[0021] Each of the reflectors 51 to 54 is made of a plate-like or sheet-like reflector having a different reflectance to the infrared rays used in the LiDAR 11. The reflectance of each reflector is, for example, 70% for the front reflector 51, 50% for the rear reflector 52, and 95% for the right reflector 53 and the left reflector 54. The reflector may be a plate-like material having a paint with a different reflectance applied to its surface, or a sheet with a different reflectance attached thereto.

[0022] Here, when there are different types of carts 4, for example when there are carts of different sizes or when there are carts with different designated storage locations, the reflectance of the four reflective materials 51-54 constituting the marker 5 of each cart is made different, or the reflectance attachment surface of each reflective material 51-54 is made different. As a result, the combination of reflectance of the four reflective materials 51-54 constituting the marker 5 becomes unique for different types of carts. In reality, there are about two to six types of carts, so it is possible to realize combinations for distinguishing these types using only the above-mentioned three reflective materials with different reflectances.

[0023] The detection flow in the cart detection system configured as above will be described. The detection here is basically a flow in which a cart present in the facility is detected, and if the detected cart remains stationary in the same position without being moved for a predetermined time or more, it is determined to be an abandoned cart. Various flows are possible for performing such detection, and one example will be described.

[0024] When the facility shown in FIG. 1 is open, the detection device 1 is constantly performing detection operations within the facility. The LiDAR 11 projects pulsed infrared light from the light projecting unit 12 while scanning, receives the reflected light reflected by the object at the light receiving unit 13, and outputs a photoelectrically converted signal. The signal processing unit 14 processes this signal using a predetermined algorithm to output a detection signal of the object. This detection signal is data on two-dimensionally arranged dots (points) obtained when the detection device 1 scans the object in the horizontal and vertical directions, and is data including information such as the distance and direction to the object and information on its light intensity. Hereinafter, the detection signal including this data is referred to as detection data. It can also be said that the light intensity information in this detection data is information on the reflectance of the surface of the object.

[0025] Fig. 6 is a detection flow diagram, and Fig. 7 is a diagram showing a schematic diagram of a part of the detection flow. The detection device 1 performs detection in a detection area, and first obtains pre-detection data as the first data (S101). Then, after a predetermined time (S102), detection is performed again to obtain post-detection data following the pre-detection data (S103). The predetermined time from the detection of the pre-detection data to the detection of the post-detection data is, for example, several tens of seconds to several minutes.

[0026] In the detection by the LiDAR 11, as described above, detection data corresponding to infrared dots scanned in a two-dimensional direction is obtained. In the upper part of Fig. 7(a), in order to simplify the explanation, only dots D in a one-dimensional direction are illustrated, and the front detection data is detected by light reflected from the cart 4 and other objects (car 201, person 202 operating another cart 4 (4A), fixed equipment (display board) 203, road cone 204, etc.). In the lower part of Fig. 7(a), the vertical axis is light intensity, and the light intensity of the detection data Dd1 detected for each dot D is shown.

[0027] The detected anterior detection data and posterior detection data are sequentially transmitted to the management control device 2. The information detection unit 22 of the management control device 2 compares the anterior detection data with the posterior detection data and extracts identical detection data (S104). If identical detection data does not exist (S105), the posterior detection data is replaced with the anterior detection data (S106), and new posterior detection data is detected again after a predetermined time has elapsed (S103). Then, the extraction of identical detection data from each of the anterior and posterior detection data is repeated as in the previous case. In this example, as shown in FIG. 7(c), detection data Dd3 of objects that do not move for a predetermined time, here the cart 4, the road cone 204, and the signboard 203, is extracted as identical detection data from the detection data Dd1 in FIG. 7(a) and the detection data Dd2 in FIG. 7(b).

[0028] When identical detection data Dd3 is extracted, the information detection unit 22 compares the identical detection data Dd3 with previously stored detection data of fixed equipment (hereinafter, fixed data), and when the same detection data as the fixed data is included, the information detection unit 22 removes the detection data (S107). Since the fixed equipment is fixedly installed in the facility, the detection data when the fixed equipment is detected prior to detection can be acquired and stored in advance in the information detection unit 22. In FIG. 7(d), the detection data identical to the fixed data of the fixed equipment (display board) 203 is removed. As a result, the remaining detection data Dd4 becomes the detection data of an object that is temporarily stationary in the facility. That is, it is the detection data of the abandoned cart 4 and the road cone 204.

[0029] Then, the cart 4 is detected based on the remaining detection signal Dd4 (S108). This detection is for the marker 5 provided on the cart 4, and the arrangement pattern of the detection data obtained when the reflectances of the four reflectors 51 to 54 constituting the marker 5 are arranged in the scanning direction, in other words, the horizontal arrangement pattern of the light intensity corresponding to each reflector is detected. For example, as shown in FIG. 8, the detection data Dd is compared with the detection thresholds Th1, Th2, and Th3 corresponding to the reflectances of the reflectors of 50%, 70%, and 95%. Then, the comparison result is binarized to obtain binary data. Here, since there are three detection thresholds, three-bit binary data of (000), (001), . . . , (100) is obtained. These binary data may be expressed as hexadecimal "0", "1", . . . "4". Then, when the marker 5 is detected, the arrangement state of each reflector 51 to 54 is detected as the arrangement state of the binary data, that is, the arrangement pattern.

[0030] Meanwhile, the information detection unit 22 detects detection data obtained by previously measuring the reflectors 51 to 54 constituting the marker 5 with the LiDAR 11 of the detection device 1, and stores the arrangement pattern of the obtained binary data. In the embodiment, when the cart 4 is detected from the right side direction, the reflectors 52, 53, and 51 are detected from left to right, which is the scanning direction. Therefore, an arrangement pattern of binary data in which the binary data corresponding to these reflectors 52, 53, and 51 are arranged in this order is stored. For example, the detection data Dd in FIG. 8 is "3", "4", "2", and "1" from the left in the case of a hexadecimal arrangement pattern.

[0031] In this case, binary data detected in different forms is obtained as binary data, and therefore a large number of binary data corresponding to various different detection forms are stored. For example, when the cart 4 is detected from various different directions, the arrangement pattern of the binary data is different. In addition, the combination of the reflective material constituting the marker 5 differs depending on the type of cart, so binary data for various markers is also stored. Furthermore, the arrangement patterns of these binary data are stored in association with the type of cart corresponding to the binary data.

[0032] Then, the information detection unit 22 compares the binary data obtained from the detected cart 4 with each arrangement pattern of the stored binary data. That is, it searches for a cart having the same arrangement pattern as the detected binary data from among the many stored arrangement patterns of binary data. This search identifies the detected object as the cart 4.

[0033] When the detection information unit 22 detects the cart 4 (S109), it detects cart information such as the position, orientation, and designated storage location of the detected cart (S110). The detected cart information is displayed on the monitor 23. A manager in the management room can go to the cart 4 based on the displayed cart information and move the cart 4 to the designated storage location. Alternatively, the management communication unit 21 can transmit the cart information to the robot communication unit 33, and in response to this, the collection robot 3 can automatically travel to the position of the cart 4, connect the arm 35 to the handle 44 of the cart 4, and move the cart 4 to the designated storage location.

[0034] In the flows of Figures 6 and 7, carts in a stationary state are extracted from the detection data detected over time, and then the cart information is detected, but carts may be detected directly. For example, although not shown, the information detection unit 22 compares all received detection data with a detection threshold value, and detects an object in the detection data with a matching array pattern as a cart. This detection is then repeated for a predetermined period of time, and when the position and direction of the detected cart do not change, it is detected as an abandoned cart. This flow increases the number of signal processing steps when the information detection unit 22 compares the array patterns of the detection data, but since the number of steps in the flow is reduced, it is effective when the signal processing capacity of the information detection unit 22 is high.

[0035] In the present invention, the reflectors 51-54 of the marker 5 are arranged on the surfaces of the basket section 43 of the cart 4 facing different directions. Therefore, depending on the relative positional relationship between the detection device 1 and the cart 4, it is difficult to detect all of the reflectors 51-54 on the four sides of the basket section 43, but it is possible to detect at least one of the right reflector 53 and the left reflector 54, and a part of the front reflector 51 and the rear reflector 52 between them. Therefore, the information detection unit 22 can perform detection with three or more different detection thresholds, thereby improving the detection accuracy. In addition, the orientation of the cart can be detected from the left-right arrangement of the reflectors with different reflectivities.

[0036] In addition, in the present invention, the detection flow of the cart based on the detection data is not specified, but in any case, the marker is composed of three or more reflecting materials with different reflectances, and the information detection unit 22 performs detection with three or more different detection thresholds. Therefore, it is possible to detect an arrangement pattern of detection data based on the reflectance of at least three reflecting materials with different reflectances. By using three reflecting materials, it is possible to diversify the arrangement pattern of a specific object. In this way, since the probability that there is another object with three different reflectances arranged in a specific order is low, it is possible to reliably detect the cart with high accuracy even if there is another object near the cart.

[0037] In the cart 4 of the embodiment shown in Fig. 5(a), the right reflective material 53 and the left reflective material 54 of the basket portion 43 constituting the marker 5 are set to the same reflectance, but they may be different. For example, the reflectance of the right reflective material 53 may be 85%, and the reflectance of the left reflective material 54 may be 95%. In this case, when detecting the cart from the detection data, it becomes possible to detect it at four detection thresholds corresponding to four different reflectances, and the detection accuracy can be further improved.

[0038] The detection device 1 and the management control device 2 may be connected to each other by local wireless communication such as a LAN. The same applies to the connection with the recovery robot 3. Alternatively, the detection device 1 and the management control device 2 may be connected by wire.

[0039] Although not shown in the figures, in the case where the detection areas of the multiple detection devices 1 shown in Fig. 1 are configured to overlap, when a cart 4 is left in this overlapping detection area, detection may be performed by combining two or three detection devices 1 whose detection areas overlap. That is, by integrating the detection data of the multiple detection devices 1 in the information detection unit 22 of the management control device 2, it is possible to detect all of the reflective materials 51-54 on the four sides of the basket part 43 of the cart 4. This increases the amount of information contained in the detection data, and further improves the detection accuracy of the cart.

[0040] In the embodiment, each of the reflectors 51 to 54 is configured as a reflecting surface with a uniform reflectance, but may be configured by combining boards, paints, and sheets with different reflectances. For example, as shown in FIG. 5(c), a reflector with a reflectance of 95% (right reflector 53) and a reflector with a reflectance of 50% (rear reflector 52) may be combined in a stripe shape, and the area ratio of these reflectors 53 and 52 may be appropriately adjusted to effectively create a reflector with a reflectance of 70%. In this example, a reflector with a reflectance of approximately 70% can be formed by forming a 95% reflector and a 50% reflector in an area ratio of approximately 1:1.

[0041] In this way, a reflector with a desired reflectance can be obtained with a reflector with a small reflectance, and the reflector can be constructed at low cost. The reflectors may be combined in a grid pattern. In either case, it is essential that the size of the stripes or grids in which the reflectors are combined be set to a dimension larger than the detection resolution of the LiDAR 11. That is, as shown in Figs. 7(a) and (b), the LiDAR 11 detects the light intensity at the dot D, so that reflectors with different reflectances are included within the diameter dimension of the dot D, and the reflectance of each reflector is averaged within this diameter dimension.

[0042] Although the embodiment is an example applied to a cart, it may also be applied to a wheelchair 6 as shown in FIG. 9. In the case of a wheelchair 6, reflective materials 55, 56, (57) are attached to the rear surface (back surface) 61 of the seat backrest, which are facing in different directions, and to both sides of the left and right armrests 62 of the seat, respectively, to form a marker 5. Reflective materials may also be attached to the seating surface of the seat and the front of the backrest. In this embodiment, by detecting the reflective materials 55, 56, (57) of the marker 5, wheelchair information can be detected, making it possible to respond appropriately.

[0043] In the embodiment, the detection device is configured with a LiDAR, but may be configured with an imaging camera. In this case, as shown in FIG. 10, the marker 5A is formed in a required array pattern by combining a plurality of rectangular cells 58 with different infrared reflectances in a grid pattern. For example, the marker 5A is configured with a combination of red cells with a reflectance of 95%, green cells with a reflectance of 70%, blue cells with a reflectance of 50%, and black cells with a reflectance of 10%. Note that each cell 58 may be configured by applying paint of a different color, or may be configured by combining sheets with different reflectances. Such a marker 5A is attached to the four side surfaces of the basket part 43 of the cart 4 as shown in FIG. 11.

[0044] Although a description of the detection device 1 that is applied when this marker 5A is used is omitted, the marker is captured by an imaging camera, and the captured image is subjected to signal processing to obtain the array pattern of the detection data. Depending on the type of imaging camera, the detection data is output as an array pattern of color images or an array pattern of monochrome images. Then, the information detection unit 22 detects the cart from the array pattern of this detection data. The detection flow at this time can be the same as that of the embodiment.

[0045] When the image captured by the imaging camera is a color image, the marker 5A is detected from the combination of colors of each cell, and the cart is detected. When the image is monochrome, the marker 5A is detected based on the shading of each cell, and the cart is detected. At this time, cart information, particularly the type of cart and the designated storage location, can be detected from the arrangement pattern and peculiarity of the cells 58 of the marker 5A. Furthermore, the position and direction of the cart can be detected by performing image analysis on the captured image of the marker.

[0046] In this embodiment, the marker 5A is also composed of a combination of cells containing three or more different reflectances, so that when detecting the array pattern, detection is performed using three or more detection thresholds, particularly image shading corresponding to three or more reflectances, thereby improving the accuracy of marker or cart detection. Furthermore, since the imaging camera is smaller and less expensive than the LiDAR, it is also effective in reducing the size and cost of the detection device or detection system.

[0047] In the present invention, detection of carts, people, vehicles, etc. may be performed using a method that uses a classifier constructed by clustering processing of point cloud data or machine learning, instead of the method described in the embodiment.

[0048] In the present invention, the management control device may be provided integrally with the detection device. In particular, in a detection system consisting of one detection device, it is preferable to configure them integrally in order to simplify the system configuration. The facility to which the present invention is applied is not limited to the shopping center of the embodiment, but can be applied to public facilities including stations and airports, and private facilities such as factories. In addition, it is not limited to outdoors, but may be constructed indoors. [Explanation of symbols]

[0049] 1. Detection device 2 Management control device 3. Recovery robot 4 Cart (Object) 5. Marker 6. Wheelchair 11 Detection unit (LiDAR) 12 Light projector 13 Light receiving part 14 Signal Processing Section 15 Detection and communication unit 21 Management and Communications Department 22 Information detection unit 23 Monitor 51~57 Reflective material 100 Designated storage area 110 Lighting 120 Management room

Claims

1. An object detection system comprising: an object having a plurality of reflective materials, each with a different reflectivity, arranged in a predetermined arrangement on a plurality of surfaces facing different directions on the object's outer surface; a detection device that detects the plurality of reflective materials with different reflectivities; and a management control device that detects the object from the arrangement state of the detected plurality of reflective materials.

2. The object detection system according to claim 1 , wherein the plurality of reflective materials includes at least three reflective materials each having a different reflectance.

3. 2. The object detection system according to claim 1, wherein the object has at least three surfaces facing in different directions, each surface being provided with a reflective material having a different reflectivity.

4. 2. The object detection system according to claim 1, wherein the reflective material is a combination of a plurality of reflective materials having different reflectances in a required area ratio, so that the desired reflectance is set.

5. The object detection system of claim 1, wherein the detection device is configured as a LiDAR that projects light onto the reflective material, receives the light reflected by the reflective material, and outputs detection data including at least information on the position and reflectivity of the reflective material.

6. 2. The object detection system according to claim 1, wherein the management control device detects each reflective material based on detection data detected by a detection device and a number of detection thresholds corresponding to the number of different reflectivities.

7. The object detection system according to claim 1 , wherein the plurality of reflective materials are composed of reflective materials having different reflectances and colors.

8. The object detection system of claim 1, wherein the object is equipment such as a cart or wheelchair rented out at a specified facility, the detection device is installed at the facility, detects the equipment within the facility and outputs the detection data, and the management control device outputs equipment information including location information of the equipment.

9. The object detection system according to claim 8 , wherein the management control device displays object information of the detected equipment on a display device.

Images