Location estimation system and location estimation method

Abstract

To provide a location estimation system and location estimation method that enable sufficient grasping of a target person's behavior.SOLUTION: A system comprises a sensing unit with multiple sensors and a detection unit that detects a target area where a target is located from multiple areas divided by boundaries of sensing areas of each of the multiple sensors on a predetermined detection surface based on a combination of sensing information indicated by the signals output from the multiple sensors. And the detection unit estimates a location of the target at a predetermined timing based on the target area at each of the multiple timings that includes the predetermined timing and one or more timings before the predetermined timing, and outputs the estimated location of the target.SELECTED DRAWING: Figure 10

Description

[Technical field]

[0001] The present invention relates to a position estimation system and a position estimation method. [Background technology]

[0002] In recent years, a service (hereinafter, also referred to as a monitoring service) has been provided in which family members living in remote locations (hereinafter, also referred to as a watcher) check the safety of elderly people (hereinafter, also referred to as a target person). In such a monitoring service, for example, a device installed in the target person's residence acquires information on the target person's behavior in real time and transmits the acquired information to a device that can be viewed by the watcher. This enables the watcher to appropriately check the safety of the target person (see Patent Documents 1 to 6). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2017-190967 A [Patent Document 2] JP 2014-169968 A [Patent Document 3] JP 2019-045279 A [Patent Document 4] JP 2011-215027 A [Patent Document 5] JP 2012-233890 A [Patent Document 6] Japanese Patent Application Publication No. 7-311280 Summary of the Invention [Problem to be solved by the invention]

[0004] Here, in the above-mentioned monitoring service, for example, from the viewpoint of protecting the privacy of the target person, a human presence sensor such as an infrared sensor (hereinafter, also simply referred to as a human presence sensor) may be used instead of a camera or the like.

[0005] However, such a human presence sensor may not be able to accurately detect the location of the target person, for example. Therefore, the above-mentioned monitoring service may not be able to accurately track the target person's behavior, for example, and may not be able to fully understand the target person's behavior.

[0006] Therefore, an object of the present invention is to provide a position estimation system and a position estimation method that make it possible to fully grasp the behavior of a target person. [Means for solving the problem]

[0007] In order to achieve the above-mentioned object, the position estimation system of the present invention comprises a sensing unit having a plurality of sensors, and a sensing unit that detects a target area in which an object is located from a plurality of areas divided by boundaries of each of the sensing areas of the plurality of sensors on a specified detection surface based on a combination of sensing information indicated by signals output from the plurality of sensors, and the sensing unit estimates a position of the object at a specified timing based on the target area for each of a plurality of timings including the specified timing and one or more timings prior to the specified timing, and outputs the estimated position of the object. Effect of the Invention

[0008] According to the position estimation system and the position estimation method of the present invention, it is possible to fully grasp the behavior of a subject. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a position estimation system 1000. [Diagram 2] FIG. 2 is a diagram showing an example of the configuration of the sensing unit 200. As shown in FIG. [Diagram 3] FIG. 3 is a diagram showing an example of the configuration of the sensing unit 200. As shown in FIG. [Figure 4] FIG. 4 is a diagram showing an example of the configuration of the sensing unit 200. As shown in FIG. [Diagram 5]FIG. 5 is a diagram illustrating the sensing areas of the plurality of human presence sensors 2. As shown in FIG. [Figure 6] FIG. 6 is a diagram illustrating the sensing areas of the plurality of human presence sensors 2. As shown in FIG. [Figure 7] FIG. 7 is a block diagram illustrating the function of the detection unit 100. As shown in FIG. [Figure 8] FIG. 8 is a flowchart illustrating the position detection process in the first embodiment. [Figure 9] FIG. 9 is a diagram illustrating a modification of the position estimation process in the first embodiment. [Figure 10] FIG. 10 is a diagram illustrating a modification of the position estimation process in the first embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the technical scope of the present invention is not limited to these embodiments.

[0011] [Example of configuration of position estimation system 1000] First, a description will be given of an example of the configuration of a position estimation system 1000. FIG.

[0012] The position estimation system 1000 includes a detection unit 100 and a sensing unit 200, as shown in FIG.

[0013] The detection unit 100 is, for example, a computer device, such as a general-purpose PC (Personal Computer). The detection unit 100 performs a process (hereinafter, also referred to as a position estimation process) of estimating the position of a target person (hereinafter, also simply referred to as a target) based on a signal detected by the sensing unit 200.

[0014] The detection unit 100 has the hardware configuration of a general-purpose computer device, and for example, as shown in Fig. 1, has a CPU 101 which is a processor, a memory 102, a communication interface 103, and a storage medium 104. Each part is connected to each other via a bus 105.

[0015] The storage medium 104 has, for example, a program storage area (not shown) that stores a program 110 for performing a position estimation process.

[0016] The storage medium 104 also has a storage area (not shown) for storing information used when performing the position estimation process, for example. The storage medium 104 may be, for example, a hard disk drive (HDD) or a solid state drive (SSD).

[0017] The CPU 101 performs the position estimation process by, for example, executing a program 110 loaded from the storage medium 104 to the memory 102 .

[0018] The communication interface 103 communicates with the sensing unit 200, for example, via wired or wireless communication.

[0019] The sensing unit 200 is, for example, a structure to which a plurality of human sensors are attached. An example of the configuration of the sensing unit 200 will be described below.

[0020] [Configuration example of sensing unit 200] Fig. 2 to Fig. 4 are diagrams showing a configuration example of the sensing unit 200. Specifically, Fig. 2 is a diagram showing a configuration example of the sensing unit 200 attached to a ceiling S (hereinafter also simply referred to as the ceiling S) of a room in a subject's residence or the like, as viewed from one horizontal direction (Y1 direction side). Also, Fig. 3 and Fig. 4 are diagrams showing a configuration example of the sensing unit 200 attached to the ceiling S, as viewed from a vertical downward direction (Z2 direction side).

[0021] 2 and 3, the sensing unit 200 has, for example, a unit body 1 and a plurality of human presence sensors 2. In the following, the sensing area of ​​each human presence sensor 2 is assumed to be rectangular. That is, in the following, the description is given assuming that a lens (not shown) attached to each human presence sensor 2 is a lens that can make the sensing area rectangular.

[0022] The unit body 1 is, for example, a flat rectangular member, and is fixed to the ceiling S by attaching one surface (the surface on the Z1 direction side) to the ceiling S. In the following, the unit body 1 will be described as being a flat rectangular member, but the unit body 1 may be, for example, a member having a shape such as a truncated cone.

[0023] Each of the plurality of human sensors 2 is, for example, a pyroelectric infrared sensor that detects a target person located in the sensing area of ​​each sensor. A pyroelectric infrared sensor is, for example, a sensor that detects a temperature change of an object, a person, etc., by utilizing a pyroelectric effect. Each of the plurality of human sensors 2 transmits a signal indicating whether or not each sensor has detected a target person to the detection unit 100 at a predetermined timing (hereinafter also simply referred to as a predetermined timing), such as every 10 milliseconds. In the following, the description will be given assuming that each of the plurality of human sensors 2 is a pyroelectric infrared sensor, but each of the plurality of human sensors 2 may be, for example, a motion detection sensor (motion sensor) such as a Doppler sensor, or a heat detection sensor.

[0024] 2, each of the plurality of human presence sensors 2 is attached, for example, to the other surface (the surface on the Z2 direction side) of the unit body 1. That is, each of the plurality of human presence sensors 2 is attached, for example, so as to face the floor surface (not shown) of the room.

[0025] More specifically, for example, six human presence sensors 2 are attached to each unit body 1 as shown in Fig. 3. Below, a case where six human presence sensors 2 are attached to the unit body 1 will be described, but for example, a number of human presence sensors 2 other than six may be attached to the unit body 1. Below, a case where six human presence sensors 2 are attached to form a regular hexagon on the XY plane will be described, but each of the six human presence sensors 2 may be attached to, for example, other locations on the unit body 1 as shown in Fig. 4. Below, a case where each of the multiple human presence sensors 2 is attached to the unit body 1 will be described, but at least some of the multiple human presence sensors 2 may be attached directly to the ceiling S without being integrated with the unit body 1, for example.

[0026] 3, the human sensor 2 attached to the X2 direction side of the unit body 1 will also be referred to as human sensor 21, the human sensor 2 attached to the Y1 direction side of the unit body 1 on the X2 direction side will also be referred to as human sensor 22, and the human sensor 2 attached to the Y1 direction side of the unit body 1 on the X1 direction side will also be referred to as human sensor 23. Also, as shown in FIG. 3, the human sensor 2 attached to the X1 direction side of the unit body 1 will also be referred to as human sensor 24, the human sensor 2 attached to the Y2 direction side of the unit body 1 on the X1 direction side will also be referred to as human sensor 25, and the human sensor 2 attached to the Y2 direction side of the unit body 1 on the X2 direction side will also be referred to as human sensor 26.

[0027] Here, each of the multiple human presence sensors 2 is attached to the unit body 1 such that the sensing areas of each human presence sensor 2 are different from each other, for example, in order to expand the overall sensing area. Furthermore, each of the multiple human presence sensors 2 is attached to the unit body 1 such that the sensing areas of each human presence sensor 2 partially overlap each other, for example, due to the need to detect the position of a target person. Specific examples of the sensing areas of the multiple human presence sensors 2 will be described below.

[0028] [Specific example of detection areas of multiple human presence sensors 2] 5 and 6 are diagrams for explaining the sensing areas of a plurality of human presence sensors 2. Specifically, Fig. 5 is a diagram showing the sensing areas of each human presence sensor 2 in an area that partially includes the sensing surface F. Also, Fig. 6 is a diagram showing the sensing areas of each human presence sensor 2 in the sensing surface F. That is, Fig. 6 is an enlarged view of the sensing surface F in Fig. 5.

[0029] The detection plane F is, for example, a plane located between the ceiling S and the floor surface in the room (a plane on the Z2 direction side of the ceiling S) and is a plane parallel to the ceiling S and the floor surface. Specifically, the detection plane F may be, for example, a plane at a height near the subject's head when the subject is standing (for example, a plane about 1.5 m away from the ceiling S) or a plane at a height near the subject's waist when the subject is standing (for example, a plane about 2.0 m away from the ceiling S).

[0030] 5 and 6, sensing area R1 is, for example, the sensing area of ​​the human presence sensor 21 described in FIG. 3, and is an area located on the opposite side (X1 direction side) of the position of the human presence sensor 21 on the XY plane when the center of the XY plane is based on the center of the detection surface F. Furthermore, sensing area R1 is, for example, an area in which a part of the outer edge (hereinafter also referred to as boundary L1) is located within the detection surface F, and the part of the outer edge other than boundary L1 is located outside the detection surface F.

[0031] 3, and is an area located on the opposite side (the side inclined 60 degrees from the X1 direction toward the Y1 direction) of the position of the human sensor 22 on the XY plane when the center of the XY plane is based on the center of the detection surface F. Furthermore, the detection area R2 is an area in which, for example, a part of the outer edge (hereinafter also referred to as boundary L2) is located within the detection surface F, and the part of the outer edge other than boundary L2 is located outside the detection surface F.

[0032] 3, and is an area located on the opposite side (the side inclined 60 degrees from the X2 direction toward the Y1 direction) of the position of the human sensor 23 on the XY plane when the center of the XY plane is based on the center of the detection surface F. Furthermore, the detection area R3 is an area in which, for example, a part of the outer edge (hereinafter also referred to as boundary L3) is located within the detection surface F, and the part of the outer edge other than boundary L3 is located outside the detection surface F.

[0033] 3, and is an area located on the opposite side (X2 direction side) of the position of the human presence sensor 24 on the XY plane when the center of the detection surface F is used as a reference. Furthermore, the detection area R4 is an area in which, for example, a part of the outer edge (hereinafter also referred to as boundary L4) is located within the detection surface F, and the part of the outer edge other than boundary L4 is located outside the detection surface F.

[0034] 3, and is an area located on the opposite side (the side inclined 60 degrees from the X2 direction toward the Y2 direction) of the position of the human presence sensor 25 on the XY plane when the center of the XY plane is based on the center of the detection surface F. Furthermore, the detection area R5 is an area in which, for example, a part of the outer edge (hereinafter also referred to as boundary L5) is located within the detection surface F, and the part of the outer edge other than boundary L5 is located outside the detection surface F.

[0035] 3, and is an area located on the opposite side (the side inclined 60 degrees from the X1 direction toward the Y2 direction) of the position of the human sensor 26 on the XY plane when the center of the XY plane is based on the center of the detection surface F. Furthermore, the detection area R6 is an area in which, for example, a part of the outer edge (hereinafter also referred to as boundary L6) is located within the detection surface F, and the part of the outer edge other than boundary L6 is located outside the detection surface F.

[0036] Hereinafter, the sensing area R1, the sensing area R2, the sensing area R3, the sensing area R4, the sensing area R5, and the sensing area R6 will be collectively referred to simply as the sensing area R. In addition, hereinafter, the boundary L1, the boundary L2, the boundary L3, the boundary L4, the boundary L5, and the boundary L6 will be collectively referred to simply as the boundary L.

[0037] Then, the boundary L1, the boundary L2, the boundary L3, the boundary L4, the boundary L5, and the boundary L6 divide the detection surface F into, for example, 19 regions D (hereinafter, also referred to as divided regions D) as shown in FIG.

[0038] Hereinafter, as shown in FIG. 6, the divided area D surrounded by the outer edge F0 of the detection surface F (hereinafter also referred to as simply the outer edge F0), boundary L1, and boundary L3 will be referred to as divided area D11, the divided area D surrounded by the outer edge F0, boundary L2, and boundary L6 will be referred to as divided area D12, the divided area D surrounded by the outer edge F0, boundary L1, and boundary L5 will be referred to as divided area D13, the divided area D surrounded by the outer edge F0, boundary L4, and boundary L6 will be referred to as divided area D14, the divided area D surrounded by the outer edge F0, boundary L3, and boundary L5 will be referred to as divided area D15, and the divided area D surrounded by the outer edge F0, boundary L2, and boundary L4 will be referred to as divided area D16.

[0039] In addition, as shown in FIG. 6, hereinafter, the divided area D surrounded by the outer edge F0, boundary L1, boundary L2, boundary L3, and boundary L6 will also be referred to as divided area D21, the divided area D surrounded by the outer edge F0, boundary L1, boundary L2, boundary L5, and boundary L6 will also be referred to as divided area D22, the divided area D surrounded by the outer edge F0, boundary L1, boundary L4, boundary L5, and boundary L6 will also be referred to as divided area D23, the divided area D surrounded by the outer edge F0, boundary L3, boundary L4, boundary L5, and boundary L6 will also be referred to as divided area D24, the divided area D surrounded by the outer edge F0, boundary L2, boundary L3, boundary L4, and boundary L5 will also be referred to as divided area D25, and the divided area D surrounded by the outer edge F0, boundary L1, boundary L2, boundary L3, and boundary L4 will also be referred to as divided area D26.

[0040] Furthermore, as shown in FIG. 6, hereinafter, the divided area D surrounded by boundaries L1, L2, and L6 will also be referred to as divided area D31, the divided area D surrounded by boundaries L1, L5, and L6 will also be referred to as divided area D32, the divided area D surrounded by boundaries L4, L5, and L6 will also be referred to as divided area D33, the divided area D surrounded by boundaries L3, L4, and L5 will also be referred to as divided area D34, the divided area D surrounded by boundaries L2, L3, and L4 will also be referred to as divided area D35, and the divided area D surrounded by boundaries L1, L2, and L3 will also be referred to as divided area D36.

[0041] Furthermore, hereinafter, as shown in FIG. 6, a divided area D surrounded by boundaries L1, L2, L3, L4, L5, and L6 will also be referred to as a divided area D41.

[0042] That is, for example, the divided area D11 is a divided area D included in the sensing area R4, the sensing area R5, and the sensing area R6, and is not included in the sensing area R1, the sensing area R2, and the sensing area R3. Therefore, for example, when the human presence sensor 24, the human presence sensor 25, and the human presence sensor 26 corresponding to the sensing area R4, the sensing area R5, and the sensing area R6, respectively, transmit a signal indicating that they have detected a target person to the detection unit 100, and when the human presence sensor 21, the human presence sensor 22, and the human presence sensor 23 corresponding to the sensing area R1, the sensing area R2, and the sensing area R3, respectively, transmit a signal indicating that they have not detected a target person to the detection unit 100, the detection unit 100 determines that the target person is located in the divided area D11.

[0043] Also, for example, the divided area D21 is a divided area D included in the sensing area R3, the sensing area R4, the sensing area R5, and the sensing area R6, and is not included in the sensing area R1 and the sensing area R2. Therefore, for example, when the human presence sensor 23, the human presence sensor 24, the human presence sensor 25, and the human presence sensor 26 corresponding to the sensing area R3, the sensing area R4, the sensing area R5, and the sensing area R6, respectively, transmit a signal indicating that they have detected a target person to the detection unit 100, and the human presence sensor 21 and the human presence sensor 22 corresponding to the sensing area R1 and the sensing area R2, respectively, transmit a signal indicating that they have not detected a target person to the detection unit 100, the detection unit 100 determines that the target person is located in the divided area D21.

[0044] Also, for example, the divided area D31 is a divided area D included in the sensing area R2, the sensing area R3, the sensing area R4, the sensing area R5, and the sensing area R6, and is not included in the sensing area R1. Therefore, for example, when the human presence sensor 22, the human presence sensor 23, the human presence sensor 24, the human presence sensor 25, and the human presence sensor 26 corresponding to the sensing area R2, the sensing area R3, the sensing area R4, the sensing area R5, and the sensing area R6 respectively transmit a signal indicating that they have detected a target person to the detection unit 100, and the human presence sensor 21 corresponding to the sensing area R1 transmits a signal indicating that they have not detected a target person to the detection unit 100, the detection unit 100 determines that the target person is located in the divided area D31.

[0045] Also, for example, the divided area D41 is a divided area D included in all of the sensing area R1, the sensing area R2, the sensing area R3, the sensing area R4, the sensing area R5, and the sensing area R6. Therefore, for example, when the human presence sensor 21, the human presence sensor 22, the human presence sensor 23, the human presence sensor 24, the human presence sensor 25, and the human presence sensor 26 corresponding to the sensing area R1, the sensing area R2, the sensing area R3, the sensing area R4, the sensing area R5, and the sensing area R6, respectively, transmits a signal indicating that the human presence sensor 21, the human presence sensor 22, the human presence sensor 23, the human presence sensor 24, the human presence sensor 25, and the human presence sensor 26, respectively, to the detection unit 100, the detection unit 100 determines that the human presence sensor 21 is located in the divided area D41.

[0046] In this way, in the position estimation system 1000, for example, by adjusting the mounting angle (azimuth angle) of each human sensor 2 in the XY plane and the mounting angle (elevation angle) of each human sensor 2 in the Z-axis direction, the position of the boundary L corresponding to each human sensor 2 is controlled and the detection surface F is divided into a plurality of divided regions D. Then, for example, when the position estimation system 1000 receives signals from each human sensor 2, it identifies a combination of contents indicated by each signal and identifies the divided region D corresponding to the identified combination as the position where the subject is present.

[0047] This enables the position estimation system 1000 to, for example, accurately identify the position where the target person is present.

[0048] In the above example, the unit body 11 is attached to the ceiling S, and the detection surface F is a plane extending on the XY plane, but this is not limiting. Specifically, the unit body 11 may be attached to, for example, a wall (wall surface) in a room. In this case, the detection surface F may be, for example, a plane extending on the XZ plane.

[0049] This allows, for example, the posture of the subject to be estimated in the position estimation system 1000. Specifically, in this case, the position estimation system 1000 can determine, for example, whether the subject is sitting or not (whether the subject is standing or not).

[0050] [Detection Unit 100 Functions] Next, a description will be given of the function of the detection unit 100. Specifically, FIG.

[0051] As shown in FIG. 7, the detection unit 100 realizes various functions including, for example, a signal receiving unit 111, an object detecting unit 112, a position estimating unit 113, and a result output unit 114.

[0052] The signal receiving section 111 of the detection unit 100 receives, for example, a signal indicating the detection result of the target person from each human presence sensor 2 attached to the sensing unit 200.

[0053] Specifically, the signal receiving unit 111 receives a signal indicating whether or not each human sensor 2 has detected a target person at each predetermined timing, for example.

[0054] The object detection section 112 of the detection unit 100 detects a subject on the detection plane F, for example, at each predetermined timing, based on a combination of information (hereinafter also referred to as detection information) indicated by a signal received by the signal receiving section 111. The detection information is, for example, information indicating that the human sensor 2 has detected a subject (hereinafter also referred to as first information), or information indicating that the human sensor 2 has not detected a subject (hereinafter also referred to as second information). That is, the object detection section 112 detects a divided area D (hereinafter also referred to as a target area D) in which a subject is located from among a plurality of divided areas D included in the detection plane F, based on a combination of information indicated by a signal received by the signal receiving section 111, for example, at each predetermined timing.

[0055] Specifically, in the example described in Figure 6, for example, when the sensing information corresponding to the signals received from each of the human presence sensors 24, 25 and 26 is first information, and the sensing information corresponding to the signals received from each of the human presence sensors 21, 22 and 23 is second information, the object detection unit 112 determines that a target is present in the divided area D11.

[0056] Furthermore, in the example described in FIG. 6, for example, when the sensing information corresponding to the signals received from each of the human presence sensors 23, 24, 25 and 26 is first information, and the sensing information corresponding to the signals received from each of the human presence sensors 21 and 22 is second information, the object detection unit 112 determines that a target is present in the divided area D21.

[0057] The position estimation section 113 of the detection unit 100 estimates the position where the subject is present (hereinafter also referred to as the estimated position) based on, for example, each target area D detected at each predetermined timing by the target detection section 112. That is, the position estimation section 113 estimates the estimated position of the subject at the most recent timing (hereinafter also referred to as the most recent timing) of the multiple predetermined timings from the position (e.g., the center position) of each target area D detected at each of the multiple predetermined timings (hereinafter also referred to as the multiple timings).

[0058] Specifically, the position estimation unit 113 estimates, for example, the average value of values ​​(e.g., coordinates on the XY plane) indicating the center position of each target area D detected at each of multiple timings as the estimated position of the target person at the most recent timing.

[0059] The result output section 114 of the detection unit 100 outputs, for example, at predetermined timings, the estimated position of the subject estimated by the position estimation section 113 to an output screen (not shown) on an output device (not shown) that can be viewed by the watcher.

[0060] Specifically, the result output unit 114 continuously outputs, for example, points indicating the X and Y coordinates of the estimated positions estimated at each of the multiple timings on the output screen. Then, the watcher watches over the subject by grasping the moving route of the subject from the trajectory of each point output on the output screen.

[0061] That is, when the center position of the target area D in which the target is present is estimated as the estimated position of the target, the estimated position of the target will change, for example, depending on the target's movement across the divided area D. Therefore, for example, when the divided area D is large or the target's movement speed is fast, the estimated position of the target at a given time may be far from the estimated position at the previous given time. Therefore, in this case, it may be difficult for the watcher to grasp the movement route of the target from each point output on the output screen, and it may be difficult to watch the target.

[0062] Therefore, for example, when estimating the position of a subject at the most recent timing, the position estimation unit 113 takes into account not only the center position of the target area D at the most recent timing, but also the center positions of each target area D at one or more timings prior to the most recent timing.

[0063] This enables the position estimation unit 113 to shorten the interval between points corresponding to the estimated positions estimated at successive times, for example. Therefore, even if the divided area D is large or the moving speed of the subject is fast, the watcher can easily identify the trajectory of each point output on the output screen by the result output unit 114, and can easily grasp the moving route of the subject.

[0064] [Flowchart of location estimation process] Next, the position estimation process will be described with reference to Fig. 8. The position estimation process in the first embodiment is shown in a flow chart.

[0065] As shown in FIG. 8, for example, the signal receiving unit 111 waits until it receives a signal indicating the result of sensing a target person from each human sensor 2 attached to the sensing unit 200 (NO in S11).

[0066] Then, when a signal indicating a result of sensing a target person is received from each human sensor 2 attached to the sensing unit 200 (YES in S11), the target detection unit 112 identifies, for example, a combination of sensing information indicated by the signal received in the process of S11 (S12). That is, the detection unit 100 executes a position estimation process, for example, every time a signal is transmitted from each human sensor 2.

[0067] Specifically, in this case, the object detection unit 112 may, for example, calculate, for each of multiple human presence sensors 2, the maximum value indicated by multiple signals output from each of the multiple human presence sensors 2 at each of multiple timings as the sensing information corresponding to each human presence sensor 2.

[0068] In other words, in the process of S12, the object detection unit 112 may, for example, correct the sensing information indicated by the signal received in the process of S11 by using each sensing information indicated by each signal received in multiple past processes of S11 (for example, multiple most recent processes of S11) rather than directly identifying the sensing information indicated by the signal received in the process of S12. Hereinafter, the sensing information after correction is also referred to as corrected sensing information. In addition, in the process of S12, the object detection unit 112 may, for example, identify a combination of corrected sensing information calculated for each of the multiple human sensors 2.

[0069] More specifically, the object detection unit 112 may calculate corrected sensing information corresponding to each human presence sensor 2 by using, for example, the following formulas (1) and (2). In the following formula (1), n ​​indicates a predetermined timing, K is a predetermined constant, x(n) indicates sensing information indicated by a signal output from each human presence sensor 2 at the predetermined timing, and x(n) with a hat indicates corrected sensing information at the predetermined timing. In the following formula (2), θ indicates a predetermined threshold value.

[0070]

number

[0071]

number

[0072] That is, for example, each human presence sensor 2 may determine that a subject does not exist in the sensing area R corresponding to each human presence sensor 2 even if a subject exists in the sensing area R corresponding to each human presence sensor 2. Also, for example, each human presence sensor 2 may determine that a subject exists in the sensing area R corresponding to each human presence sensor 2 even if a subject does not exist in the sensing area R corresponding to each human presence sensor 2. Therefore, for example, the detection unit 100 may not be able to accurately identify the divided area D in which the subject exists.

[0073] Therefore, in the process of S12, the detection unit 100 in this embodiment calculates corrected detection information by using each piece of detection information indicated by the signal received in the process of S11 performed multiple times in the past. Then, the detection unit 100 performs the process of S13 and thereafter by using the calculated corrected detection information.

[0074] As a result, the detection unit 100 in this embodiment is able to accurately identify the divided area D in which the subject is present, even if, for example, each human presence sensor 2 is unable to accurately detect the subject.

[0075] In addition, in the processing of S12, the target detection unit 112 may, for example, calculate, for each human presence sensor 2, the average value of the values ​​indicated by multiple signals output from each human presence sensor 2 at each of multiple specified timings as the sensing information corresponding to each human presence sensor 2.

[0076] Specifically, the object detection unit 112 may use, for example, the following formula (3) to calculate the corrected sensing information corresponding to each human sensor 2. In the following formula (3), m is a variable that varies from 0 to K.

[0077]

number

[0078] Note that K in the above formulas (1) to (3) may be, for example, 10. That is, for example, when the predetermined timing is every 20 milliseconds, the detection unit 100 may calculate corrected sensing information corresponding to each of the plurality of human presence sensors 2 by using a signal output from each human presence sensor 2 for 200 milliseconds.

[0079] Returning to FIG. 8, the object detection unit 112 identifies an object area D in which the object is located from among a plurality of divided areas D in the detection surface F, based on, for example, the combination of sensing information identified in the process of S12 (S13).

[0080] Specifically, for example, in the example described in Figure 6, when the sensing information corresponding to the signals received from each of the human presence sensors 24, 25, and 26 is first information, and the sensing information corresponding to the signals received from each of the human presence sensors 21, 22, and 23 is second information, the target detection unit 112 identifies the divided area D11 as the target area D.

[0081] Also, for example, in the example described in Figure 6, when the sensing information corresponding to the signals received from each of the human presence sensors 23, 24, 25 and 26 is first information, and the sensing information corresponding to the signals received from each of the human presence sensors 21 and 22 is second information, the target detection unit 112 identifies the divided area D21 as the target area D.

[0082] Thereafter, the position estimation unit 113 estimates the estimated position of the target person based on, for example, the center position of the target area D identified in the process of S13 (S14).

[0083] Specifically, the position estimation unit 113 estimates, for example, the average value of values ​​(e.g., coordinates on the XY plane) indicating the center position of each target area D detected at each of multiple timings as the estimated position of the target person at the most recent timing.

[0084] In other words, in the processing of S14, the position estimation unit 113 may calculate the estimated position of the subject by using the center positions of each target area D identified in multiple past processings of S13 (e.g., multiple most recent processings of S13) rather than directly using the center position of the target area D identified in the processing of S13 as the estimated position of the subject.

[0085] More specifically, the position estimation unit 113 may calculate the estimated position of the subject by using, for example, the following formula (4): In the following formula (4), n indicates a predetermined timing, L is a predetermined constant, m is a variable that varies from 0 to L, c(n) indicates the center position of the target area D (the target area D identified in the process of S13) identified using the signals output from each human presence sensor 2 at the predetermined timing, and c(n) with a hat indicates the estimated position of the subject corresponding to the predetermined timing.

[0086]

number

[0087] This allows the detection unit 100 in this embodiment to, for example, shorten the interval between points corresponding to the estimated positions estimated at each successive timing, which allows the watcher to, for example, easily identify the trajectory of each point output on the output screen by the result output unit 114, and easily grasp the movement route of the target person.

[0088] In addition, the hatched c(nm) in formula (4) may include, for example, a weight W m may be multiplied.

[0089]

number

[0090] Specifically, W in the above formula (5)m For example, W may be a variable that varies depending on the value of m. More specifically, m may be, for example, a variable that takes on a smaller value as the value of m increases.

[0091] That is, in the position estimation process, the weight multiplied by the value calculated based on the signal output from the human presence sensor 2 at a first timing may be greater than the weight multiplied by the value calculated based on the signal output from the human presence sensor 2 at a second timing that is earlier than the first timing.

[0092] This allows the detection unit 100 in this embodiment to make adjustments so that, for example, the more recent the output timing of a signal from the human presence sensor 2, the greater the influence it has on the estimation of the estimated position of the subject. Therefore, the detection unit 100 can, for example, estimate the estimated position of the subject while more closely reflecting the actual behavior of the subject.

[0093] In addition, W in the above formula (5) m may be a variable that varies depending on the divided area D including the estimated position indicated by the hatched c(nm). Specifically, the weight corresponding to the divided area D in which the probability of the subject's presence is equal to or greater than a predetermined value (hereinafter also referred to as the first divided area D) may be greater than the weight corresponding to the divided area D in which the probability of the subject's presence is less than a predetermined value (hereinafter also referred to as the second divided area D). The first divided area D may be, for example, a divided area D showing a bed (in other words, a divided area D in which it is possible to determine that the subject may be present). The second divided area D may be, for example, a divided area D showing a desk or a shelf (in other words, a divided area D in which it is possible to determine that the possibility of the subject being present is extremely low).

[0094] That is, in the position estimation process, the weight multiplied to the value calculated based on the signal output at the time when the subject was present in the first divided area D may be greater than the weight multiplied to the value calculated based on the signal output at the time when the subject was present in the second divided area D, for example.

[0095] Thereby, the detection unit 100 in this embodiment can suppress the influence on the estimation of the estimated position of the subject caused by the sensing information that can determine that the detection of the subject by the human presence sensor 2 was not performed with high accuracy, for example. Therefore, the position estimation unit 113 can accurately estimate the estimated position of the subject even if the detection of the subject by the human presence sensor 2 was not performed with high accuracy, for example.

[0096] In addition, L in the above formula (4) and formula (5) may be, for example, 10. That is, for example, when the predetermined timing is a timing of every 20 milliseconds, the detection unit 100 may estimate the estimated position of the subject by using the center position of each target region D identified during 200 milliseconds.

[0097] Returning to FIG. 8, the result output unit 114 outputs, for example, the estimation result of the subject estimated in the process of S14 to an output screen of an output device viewable by the watcher (S15).

[0098] Thus, the position estimation system 1000 in this embodiment includes a sensing unit 200 having, for example, a unit body 1 and a plurality of human sensors 2 attached to the unit body 1 to sense a target person. The position estimation system 1000 also includes a sensing unit 100 that detects a target area D in which a target person is located from a plurality of divided areas D divided by the boundaries of the sensing areas R of the plurality of human sensors 2 on a predetermined sensing surface F based on a combination of sensing information indicated by signals output from the plurality of human sensors 2. The sensing unit 100 estimates an estimated position of the target person at a predetermined timing based on the target area D for each of a plurality of predetermined timings including, for example, a predetermined timing and one or more timings before the predetermined timing, and outputs the estimated position of the target person.

[0099] This allows the detection unit 100 in this embodiment to, for example, shorten the interval between points corresponding to the estimated positions estimated at each successive timing, which allows the watcher to, for example, easily identify the trajectory of each point output on the output screen by the result output unit 114, and easily grasp the movement route of the target person.

[0100] In the above example, the detection unit 100 and the sensing unit 200 are separate, but the present invention is not limited to this. Specifically, the detection unit 100 and the sensing unit 200 may be, for example, a single device. That is, the detection unit 100 and the sensing unit 200 may be, for example, attached to the ceiling S in an integrated state.

[0101] [Modification of the position estimation process in the first embodiment] 9 and 10 are diagrams for explaining a modified example (hereinafter also referred to as a first modified example) of the position estimation process in the first embodiment. In the following, the description will be given on the assumption that the first information in the sensing information is "1" and the second information in the sensing information is "0".

[0102] As described above, for example, each human presence sensor 2 may not be able to detect a target person even if the target person is present in the sensing area R corresponding to each human presence sensor 2. In this case, for example, the detection unit 100 may not be able to accurately identify the divided area D in which the target person is present.

[0103] Therefore, when the number of first information (sensing information indicating that a target person has been detected) included in the combination of sensing information identified in the process of S12 is less than a predetermined number, the detection unit 100 may identify a human presence sensor 2 (hereinafter also referred to as a specific human presence sensor 2) that is attached to a position adjacent to the human presence sensor 2 that outputs a signal corresponding to the first information and that outputs a signal corresponding to the second information (sensing information indicating that a target person has not been detected) among the human presence sensors 2. Then, the detection unit 100 may change the second information indicated by the signal output from the specific human presence sensor 2 among the second information included in the combination of sensing information identified in the process of S12 to the first information, and then perform the processes from S13 onward.

[0104] In other words, for example, if the number of first information included in the combination of sensing information identified in the processing of S12 is less than a predetermined number, the detection unit 100 may determine that there is a human presence sensor 2 that has failed to detect the subject, and change the sensing information indicated by the signal output from the human presence sensor 2 that can be determined to have failed to detect the subject.

[0105] Specifically, as shown in the sensing information DT11 of FIG. 9(A), when the value (fifth value) corresponding to human presence sensor 25 is "1" and the values ​​(first, second, third, fourth and sixth values) corresponding to each of human presence sensors 21, 22, 23, 24 and 26 are "0", the detection unit 100 may, for example, change the values ​​(fourth and sixth values) corresponding to each of human presence sensors 24 and 26 adjacent to human presence sensor 2 (human presence sensor 25) corresponding to "1" from "0" to "1" as shown in the sensing information DT12 of FIG. 9(A).

[0106] Furthermore, as shown in the detection information DT21 of FIG. 9(B), when the values ​​(third and fourth values) corresponding to each of the human presence sensors 23 and 24 are "1" and the values ​​(first, second, fifth and sixth values) corresponding to each of the human presence sensors 21, 22, 25 and 26 are "0", the detection unit 100 may, for example, change the values ​​(second and fifth values) corresponding to each of the human presence sensors 22 and 25 adjacent to the human presence sensor 2 (human presence sensor 23 and human presence sensor 24) corresponding to "1" from "0" to "1" as shown in the detection information DT22 of FIG. 9(B).

[0107] Also, as shown in the sensing information DT31 in Fig. 9(C), when the values ​​(fifth and sixth values) corresponding to the human sensor 25 and the human sensor 26 are "1" and the values ​​(first, second, third and fourth values) corresponding to the human sensor 21, the human sensor 22, the human sensor 23 and the human sensor 24 are "0", the detection unit 100 may change the values ​​(first and fourth values) corresponding to the human sensor 21 and the human sensor 24 adjacent to the human sensor 25 and the human sensor 26 corresponding to "1" from "0" to "1", for example, as shown in the sensing information DT32 in Fig. 9(C). A specific example of the first modified example will be described below.

[0108] [Specific example of the first modified example] FIG. 10 is a diagram illustrating a specific example of the first modified example.

[0109] For example, in the example described in FIG. 6, when the target person is located in the divided region D13, the human presence sensor 22, the human presence sensor 23, and the human presence sensor 24 each transmit a signal corresponding to the first information to the detection unit 100.

[0110] However, in this case, for example, if the human presence sensor 22 and the human presence sensor 24 cannot detect the target person, each of the human presence sensor 22 and the human presence sensor 24 may transmit a signal corresponding to the second information to the detection unit 100. In this case, the detection unit 100 may estimate, for example, a point P1 outside the detection surface F as the estimated position of the target person, as shown in Fig. 10.

[0111] Therefore, for example, after the process of S12 is performed, the detection unit 100 determines whether the number of pieces of first information included in the combination of sensing information identified in the process of S12 is a predetermined number (for example, 3) or more. If it is determined that the number of pieces of first information included in the combination of sensing information identified in the process of S12 is not the predetermined number or more, the detection unit 100 performs the process from S13 onward, assuming that signals corresponding to the first information have been transmitted from the human presence sensors 22 and 24 adjacent to the human presence sensor 23 that output a signal corresponding to the first information. That is, in this case, the detection unit 100 can estimate, for example, point P2 inside the detection surface F (point P2 included in the divided region D13) as the estimated position of the target person.

[0112] As a result, the detection unit 100 in this modified example is able to accurately identify the movement route of the target person even in a case where, for example, each human presence sensor 2 cannot accurately detect the target person. [Explanation of symbols]

[0113] 1: Unit body 2: Human presence sensor 100: Detection unit 101:CPU 102: Memory 103: Communication interface 104:Storage medium 105: Bus 111: Signal receiving unit 112: Target detection unit 113:Position estimation part 114: Result output section 200:Sensing unit

Claims

1. A sensing unit having a plurality of sensors; a detection unit configured to detect a target area in which a target is located from a plurality of areas divided by boundaries of the sensing areas of the plurality of sensors on a predetermined sensing surface based on a combination of sensing information represented by signals output from the plurality of sensors; The detection unit includes: estimating a position of the target at a predetermined time based on the target region for each of a plurality of times including a predetermined time and one or more times before the predetermined time; outputting the estimated position of the object; A position estimation system comprising:

2. Each of the sensing unit and the detection unit is a separate unit. The position estimation system according to claim 1 .

3. The sensing unit and the detection unit are an integrated unit.

2. The position estimation system according to claim 1,

4. The detection unit estimates an average value of values ​​indicating the position of the target region for each of the plurality of timings as the position of the target at the predetermined timing. The position estimation system according to claim 1 .

5. The detection unit includes: calculating a first value for each of the plurality of timings by multiplying a weight corresponding to each timing by a value indicating a position of the target region corresponding to each timing; an average value of the first values ​​for each of the plurality of timings is estimated as the position of the target at the predetermined timing; 5. The position estimation system according to claim 4.

6. the weight corresponding to a first timing included in the plurality of timings is a value greater than the weight corresponding to a second timing included in the plurality of timings and preceding the first timing; The position estimation system according to claim 5 .

7. The detection unit includes: calculating a second value by multiplying a weight corresponding to the target region at each of the plurality of timings by a value indicating a position of the target region corresponding to each of the timings; an average value of the second values ​​for each of the plurality of timings is estimated as the position of the target at the predetermined timing; 5. The position estimation system according to claim 4.

8. The weight corresponding to a first region included in the plurality of regions is a value larger than the weight corresponding to a second region included in the plurality of regions and having a lower probability of the object being located therein than the first region. The position estimation system according to claim 7 .

9. The detection unit calculates, for each of the plurality of sensors, an average value of values ​​indicated by a plurality of signals output from each of the plurality of sensors at a plurality of timings as the sensing information corresponding to each sensor. The position estimation system according to claim 1 .

10. The detection unit calculates, for each of the plurality of sensors, a maximum value indicated by a plurality of signals output from each of the plurality of sensors at a plurality of timings as the sensing information corresponding to each sensor. The position estimation system according to claim 1 .

11. The sensing information indicates either first information indicating that the object has been sensed or second information indicating that the object has not been sensed, The detection unit includes: When the number of the first information included in the combination of the sensing information is less than a predetermined number, a specific sensor that is attached to a position adjacent to a sensor that outputs a signal corresponding to the first information and that outputs a signal corresponding to the second information is identified among the plurality of sensors; changing information indicated by a signal output from the specific sensor from the combination of the sensing information from the second information to the first information; The position estimation system according to claim 1 .

12. A position estimation method for a position estimation system having a sensing unit having a plurality of sensors, and a detection unit that detects a target area in which a target is located from a plurality of areas divided by boundaries of the sensing areas of the plurality of sensors on a predetermined sensing surface based on a combination of sensing information indicated by signals output from the plurality of sensors, comprising: The detection unit includes: estimating a position of the target at a predetermined time based on the target region for each of a plurality of times including a predetermined time and one or more times before the predetermined time; outputting the estimated position of the object; A position estimation method comprising:

Images