Object detection system

The object detection system addresses the challenge of weak floor vibrations by employing multiple vibration application and detection mechanisms, enabling accurate estimation of object presence, position, and type on a floor.

WO2026235170A1PCT designated stage Publication Date: 2026-11-12PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2026/017505
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-07
Filing Date
2026-05-07
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Existing object detection systems using vibration sensors struggle with accurately detecting the presence, position, and type of objects on a floor due to weak floor vibrations, leading to improper detection.

Method used

An object detection system utilizing a configuration of multiple vibration application mechanisms and detection mechanisms, including piezoelectric elements, to apply and detect vibrations, with a control device estimating the presence, position, and type of objects based on these detections.

Benefits of technology

The system effectively estimates the presence, position, and type of objects on a floor with high accuracy using a simple configuration, improving detection reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2026017505_12112026_PF_FP_ABST
    Figure JP2026017505_12112026_PF_FP_ABST
Patent Text Reader

Abstract

The present disclosure provides an object detection system that has a simple configuration and appropriately estimates at least one among the presence or absence of an object placed on a floor, the position of the object, and the type of the object. The object detection system comprises: a plurality of vibration application mechanisms that apply vibrations to at least one among the floor and a wall; a plurality of vibration detection mechanisms that detect vibrations propagating through at least one among the floor and the wall; and a control device that controls the plurality of vibration application mechanisms. The control device comprises: an estimation unit that estimates at least one among the presence or absence of an object placed on the floor, the position of the object, and the type of the object on the basis of the detection results of the plurality of vibration detection mechanisms; and a notification unit that notifies the estimation results of the estimation unit.
Need to check novelty before this filing date? Find Prior Art

Description

Object detection system

[0001] The present disclosure relates to an object detection system.

[0002] Patent Document 1 discloses a detection system that monitors a target area based on detection results from a plurality of detection devices. As the detection devices, for example, a camera that detects movement of an object as an event, a vibration sensor that detects vibration as an event, and the like are used.

[0003] Japanese Unexamined Patent Application Publication No. 2023-131634

[0004] The present disclosure provides an object detection system that appropriately estimates at least one of the presence or absence of an object placed on a floor, the position of the object, and the type of the object with a simple configuration.

[0005] This specification includes all the contents of Japanese Patent Application No. 2025-077483 filed on May 7, 2025. The object detection system according to the present disclosure includes a plurality of vibration application mechanisms that apply vibration to at least one of a floor and a wall, a plurality of vibration detection mechanisms that detect vibration propagating through at least one of the floor and the wall, and a control device that controls the plurality of vibration application mechanisms, wherein the control device includes an estimation unit that estimates at least one of the presence or absence of an object placed on the floor, the position of the object, and the type of the object based on detection results from the plurality of vibration detection mechanisms, and a notification unit that notifies an estimation result of the estimation unit.

[0006] According to the object detection system of the present disclosure, at least one of the presence or absence of an object placed on a floor, the position of the object, and the type of the object can be appropriately estimated with a simple configuration.

[0007] FIG. 1 is a diagram illustrating an example of the configuration of the object detection system according to the embodiment. FIG. 2 is a diagram illustrating an example of the arrangement of each device of the object detection system according to the embodiment. FIG. 3 is a diagram illustrating an example of the configuration of each device of the object detection system according to the embodiment. FIG. 4 is a diagram illustrating an example of the configuration of a first trained model. FIG. 5 is a diagram illustrating an example of the configuration of a second trained model. FIG. 6 is a flowchart illustrating an example of processing of the control device

[0008] (Knowledge and other information forming the basis of this disclosure) At the time the inventors conceived of this disclosure, as shown in Patent Document 1, there was a detection system that detected objects in a target area based on the detection results of multiple vibration sensors. However, with the above technology, there were cases where objects could not be properly detected by the vibration sensors because, for example, the vibration of the floor was weak. The inventors discovered this problem and came to form the subject of this disclosure in order to solve it. Therefore, this disclosure provides an object detection system that, with a simple configuration, can properly estimate at least one of the presence or absence of an object placed on the floor, the position of the object, and the type of the object.

[0009] The embodiments will be described in detail below with reference to the drawings. However, some descriptions may be omitted to avoid unnecessary detail. For example, detailed explanations of already well-known matters or redundant explanations of substantially identical configurations may be omitted. The accompanying drawings and the following description are provided to enable those skilled in the art to fully understand this disclosure and are not intended to limit the subject matter described in the claims.

[0010] (Embodiment 1) [1-1-1. Configuration of the Object Detection System] First, the configuration of the object detection system 1 will be described with reference to Figure 1. Figure 1 is a diagram showing an example of the configuration of the object detection system 1 according to Embodiment 1. As shown in Figure 1, the object detection system 1 comprises a control device 4, a vibrator 6, a sensor 7, and a smartphone 8. The control device 4 is connected to the vibrator 6, the sensor 7, and the smartphone 8 in a communication manner.

[0011] The vibrator 6 comprises a plurality of vibrators that apply vibration to at least one of the floor and the wall. The vibrator 6 comprises, for example, a first vibrator 24 and a second vibrator 34C. In Embodiment 1, for example, the first vibrator 24 and the second vibrator 34C apply vibration to the wall, as will be explained with reference to Figure 2. Each of the first vibrator 24 and the second vibrator 34C is composed of, for example, a piezoelectric element. The first vibrator 24 is positioned on the switch 20, as shown in Figure 2. The switch 20 is positioned on the wall 94. The first vibrator 24 applies vibration to the wall 94 in a direction perpendicular to it (= Y-axis direction). The vibration applied from the first vibrator 24 to the wall 94 propagates radially through the wall 94 from the first vibrator 24 and also propagates to adjacent walls 95, 93, the ceiling 92, and the floor 91. The second vibrator 34C is positioned in the outlet 30C, as shown in Figure 2. The outlet 30C is positioned in the wall 93. The second vibrator 34C applies vibration perpendicular to the wall 93 (in the X-axis direction). The vibration applied to the wall 93 from the second vibrator 34C propagates radially through the wall 93 from the second vibrator 34C, and also propagates to adjacent walls 94, 96, the ceiling 92, and the floor 91. The first vibrator 24 and the second vibrator 34C each correspond to an example of a "vibration application mechanism." The vibrator 6 corresponds to an example of a "multiple vibration application mechanisms." The first vibrator 24 and the second vibrator 34C will be further explained with reference to Figures 2 and 3.

[0012] Embodiment 1 describes a case where the "multiple vibration application mechanisms" consist of a first vibrator 24 and a second vibrator 34C, but the embodiments are not limited to this. The "multiple vibration application mechanisms" may include three or more vibrators.

[0013] Embodiment 1 describes a case in which the first vibrator 24 and the second vibrator 34C are each composed of piezoelectric elements, but the embodiments are not limited to this. The first vibrator 24 and the second vibrator 34C may each be composed of a vibration motor, an electrostrictive oscillator, or a magnetostrictive oscillator. A vibration motor generates vibrations by using centrifugal force, for example, by attaching an eccentric weight to the rotating shaft of a motor and rotating it. An electrostrictive oscillator is an element that uses electrical strain to convert electrical vibrations into mechanical vibrations. Electrical strain is a phenomenon in which a distortion proportional to the square of the electric field occurs when an electric field is applied to a transducer. A magnetostrictive oscillator is an element that uses the magnetostrictive effect of a magnetic material to convert electrical vibrations into mechanical vibrations. The magnetostrictive effect is the effect in which a magnetic material expands and contracts due to a magnetic field.

[0014] Sensor 7 includes a plurality of vibration sensors that detect vibrations propagating through at least one of the floor and the wall. Sensor 7 includes, for example, a first sensor 34A and a second sensor 34B. Each of the first sensor 34A and the second sensor 34B is composed of a so-called "vibration sensor". A vibration sensor is composed of, for example, a displacement sensor or an acceleration sensor and detects the magnitude (displacement, velocity, or acceleration) and period (frequency) of vibration. In Embodiment 1, for example, each of the first sensor 34A and the second sensor 34B detects vibrations of the wall, as will be explained with reference to Figure 2. The first sensor 34A is located on the outlet 30A, as shown in Figure 2. The outlet 30A is located on the wall 94. The first sensor 34A detects vibrations perpendicular to the wall 94 (= Y-axis direction). The second sensor 34B is located on the outlet 30B, as shown in Figure 2. The outlet 30B is located on the wall 95. The second sensor 34B detects vibrations perpendicular to the wall 95 (i.e., in the X-axis direction). The first sensor 34A and the second sensor 34B each correspond to an example of a "vibration detection mechanism". Sensor 7 corresponds to an example of a "multiple vibration detection mechanisms". The first sensor 34A and the second sensor 34B will be further explained with reference to Figures 2 and 3.

[0015] Embodiment 1 describes a case where the "multiple vibration detection mechanisms" consist of a first sensor 34A and a second sensor 34B, but the embodiments are not limited thereto. The "multiple vibration detection mechanisms" may include three or more vibration detection sensors.

[0016] Embodiment 1 describes a case where each of the "multiple vibration detection mechanisms" is composed of a "vibration sensor," but the embodiments are not limited to this. Each of the "multiple vibration detection mechanisms" may be composed of, for example, a microphone.

[0017] Based on the detection results of the first sensor 34A and the second sensor 34B, the control device 4 estimates at least one of the presence or absence of an object placed on the floor, the position of the object, and the type of the object. The control device 4 also notifies the smartphone 8 of the estimation result. Embodiment 1 describes the case where the "object" is a desk DK and a human body P. The desk DK and the human body P are located on the floor 91, which will be described with reference to Figure 2. Embodiment 1 also describes the case where the "position of the object" is estimated. Position XD is the X coordinate of the position of the desk DK, and position YD is the Y coordinate of the position of the desk DK. Position XP is the X coordinate of the position of the human body P, and position YP is the Y coordinate of the position of the human body P. If the "object" is the desk DK, the control device 4 notifies the smartphone 8 of position XD and position YD. If the "object" is the human body P, the control device 4 notifies the smartphone 8 of position XP and position YP. The control device 4 communicates wirelessly with the smartphone 8, for example, in accordance with the Wi-Fi® standard. Alternatively, the control device 4 may communicate wirelessly with the smartphone 8, for example, in accordance with the Bluetooth® standard.

[0018] In Embodiment 1, the control device 4 estimates the position of an object placed on the floor 91 based on the detection results of the first sensor 34A and the second sensor 34B, but the embodiments are not limited to this. The control device 4 only needs to estimate at least one of the presence or absence of an object placed on the floor, the position of the object, and the type of the object, based on the detection results of the first sensor 34A and the second sensor 34B. For example, the control device 4 may detect the presence or absence of an object based on the detection results of the first sensor 34A and the second sensor 34B. Alternatively, the control device 4 may detect the type of object based on the detection results of the first sensor 34A and the second sensor 34B.

[0019] Embodiment 1 describes the case where the "object" is a desk DK and a human body P, but the embodiment is not limited to this. The "object" can be any object placed on the floor 91 shown in Figure 2. The "object" may be, for example, a chair, a bookshelf, a bed, or a sofa. The "object" may also be, for example, a bag, a shopping bag, etc.

[0020] Smartphone 8 is, for example, a smartphone owned by the user corresponding to human body P. When smartphone 8 receives position XD and position YD from control device 4, it displays the following text information as a pop-up: "The desk's position is position XD = AAAA, position YD = BBBB." "AAA" is an example of a position XD value. "BBBB" is an example of a position YD value. Also, when smartphone 8 receives position XP and position YP from control device 4, it displays the following text information as a pop-up: "The human body's position is position XP = CCCC, position YP = DDDD." "CCCC" is an example of a position XP value. "DDDD" is an example of a position YP value. Position XD, position YD, position XP, and position YP correspond to an example of an "estimated result."

[0021] Embodiment 1 describes a case in which the control device 4 notifies the smartphone 8 of the estimation result, but the embodiments are not limited to this. The control device 4 may also notify a remote control that is configured to communicate with the control device 4 of the estimation result.

[0022] [1-1-2. Configuration of the Vibrator and Vibration Sensor] Next, the arrangement of each device in the object detection system 1 will be described with reference to Figure 2. Figure 2 is a diagram showing an example of the arrangement of each device in the object detection system 1. The living space 9 shown in Figure 2 is a space enclosed by a floor 91, a ceiling 92, and four walls 93-96. A door is provided in the living space 9, and when the door is closed, the living space 9 becomes a closed space.

[0023] Figure 2 shows the X, Y, and Z axes. The X, Y, and Z axes are virtual axes that are orthogonal to each other, with the Z axis corresponding to the height direction of the living space 9. The X axis corresponds to the width direction of the living space 9, i.e., the left-right direction. The Y axis corresponds to the depth direction of the living space 9, i.e., the front-back direction. The positive direction of the Z axis is upward. The positive direction of the X axis is to the left. The positive direction of the Y axis is backward.

[0024] A vibrator 6 and a sensor 7 are placed in the living space 9. The multiple vibrators that make up the vibrator 6 and the multiple vibration sensors that make up the sensor 7 can be installed on the floor 91 and the four walls 93-96 of the living space 9. In Embodiment 1, each of the multiple vibrators that make up the vibrator 6 and the multiple vibration sensors that make up the sensor 7 are incorporated into "electrical equipment" installed in the living space 9.

[0025] "Electrical equipment" includes components, equipment, machinery, and appliances that are connected to the commercial power supply and permanently installed in the living space. More specifically, "electrical equipment" includes switches, outlets, and other power wiring components. It also includes ventilation fans, lighting fixtures, intercoms, and other power-consuming devices. Furthermore, it includes distribution boards, building energy management devices, and other power control equipment.

[0026] For example, the living space 9 shown in Figure 2 includes a lighting device 10, a switch 20, and outlets 30A, 30B, and 30C as "electrical equipment." The "electrical equipment" may also include a distribution board 50 installed inside or outside the living space 9.

[0027] Power lines 5A, 5B, 5C, and 5D are drawn into living space 9 from the distribution board 50. Power lines 5A, 5B, 5C, and 5D are, for example, wires laid concealed within the walls of living space 9, and are power cables such as VVF cables and CV cables. In the following explanation, when power lines 5A, 5B, 5C, and 5D are not distinguished from each other, they may be referred to as power line 5.

[0028] The lighting device 10 is attached to the ceiling 92. The lighting device 10 may be directly fixed to the ceiling 92, or it may be attached to a lighting duct rail or the like fixed to the ceiling 92. In Embodiment 1, a ceiling socket 112 is fixed to the ceiling 92, and the lighting device 10 is attached to the ceiling socket 112. The ceiling socket 112 may be of a form called a lighting rosette. The ceiling socket 112 corresponds to an example of a fixing member. The ceiling socket 112 is connected to a power line 5A, and the lighting device 10 is connected to the power line 5A via the ceiling socket 112.

[0029] The lighting device 10 comprises a light source 11, which will be described later with reference to Figure 3, and a cover 110 that diffuses the light emitted by the light source 11. The light source 11 is provided in the circuit body 111 together with a driver 12, etc., which will be described later with reference to Figure 3, and is covered by the cover 110.

[0030] The switch 20 has a switch body 21 that is connected to the power line 5A. The switch 20 connects and disconnects the power line 5A between the lighting device 10 and the distribution board 50. When the switch 20 is ON, power is supplied to the lighting device 10. When the switch 20 is OFF, the power supply to the lighting device 10 is cut off, and the lighting device 10 turns off.

[0031] Outlets 30A, 30B, and 30C are power outlets (also called outlets) to which load equipment that operates using power supplied from the commercial power supply can be connected. For example, outlets 30A, 30B, and 30C are single-phase 100V power outlets, with outlet 30A connected to the distribution board 50 by power line 5B, outlet 30B connected to the distribution board 50 by power line 5C, and outlet 30C connected to the distribution board 50 by power line 5D.

[0032] The distribution board 50 branches off the service drop line 58, which is connected to the commercial power system, and connects it to the power line 5. The distribution board 50 is fixed to the wall inside or outside the living space 9.

[0033] A vibrator 6 and a sensor 7 are placed among the electrical equipment installed in the living space 9.

[0034] The switch 20 includes a first vibrator 24. The first vibrator 24 applies vibration to the wall 94, for example, in a notch in the wall 94 formed for the installation of the switch 20. The first vibrator 24 applies vibration at a first frequency FQ1 or a second frequency FQ2 according to instructions from the control device 4. The first vibrator 24 includes a vibration sensor (not shown) that detects the vibration applied by the first vibrator 24 to the wall 94. The vibration sensor (not shown) outputs the detected applied vibration signal VA1 to the control device 4. When the first vibrator 24 applies vibration at the first frequency FQ1, the vibration sensor (not shown) outputs the detected applied vibration signal VA11 to the control device 4. When the first vibrator 24 applies vibration at the second frequency FQ2, the vibration sensor (not shown) outputs the detected applied vibration signal VA12 to the control device 4.

[0035] The outlet 30C is equipped with a second vibrator 34C. The second vibrator 34C applies vibration to the wall 93, for example, in a notch in the wall 93 formed for the outlet 30C. The second vibrator 34C applies vibration of a first frequency FQ1 or a second frequency FQ2 according to instructions from the control device 4. The second vibrator 34C is equipped with a vibration sensor (not shown) that detects the vibration applied by the second vibrator 34C to the wall 94. The vibration sensor (not shown) outputs the detected applied vibration signal VA2 to the control device 4. When the second vibrator 34C applies vibration of the first frequency FQ1, the vibration sensor (not shown) outputs the detected applied vibration signal VA21 to the control device 4. When the second vibrator 34C applies vibration of the second frequency FQ2, the vibration sensor (not shown) outputs the detected applied vibration signal VA22 to the control device 4.

[0036] Outlet 30A is equipped with a first sensor 34A, and outlet 30B is equipped with a second sensor 34B. The first sensor 34A detects vibrations of the wall 94, for example, in a notch in the wall 94 formed for the placement of outlet 30A. The second sensor 34B detects vibrations of the wall 95, for example, in a notch in the wall 95 formed for the placement of outlet 30B.

[0037] The dashed line, representing the first path PH1, shows an example of a path through which vibrations applied to the wall 94 by the first vibrator 24 reach the first sensor 34A via the floor 91. As shown in the first path PH1, vibrations applied to the wall 94 by the first vibrator 24 propagate, for example, downward (= negative Z-axis direction), propagate to the right (= negative X-axis direction) across the floor 91, and propagate upward (= positive Z-axis direction) across the wall 94 to reach the first sensor 34A. The dashed line, representing the second path PH2, shows an example of a path through which vibrations applied to the wall 94 by the first vibrator 24 reach the second sensor 34B via the floor 91. As shown in the second path PH2, the vibration applied to the wall 94 by the first vibrator 24 propagates, for example, downward (= negative Z-axis direction), propagates through the floor 91 to the left front, and propagates through the wall 94 upward (= positive Z-axis direction) to reach the second sensor 34B.

[0038] The dashed line, representing the third path PH3, shows an example of a path through which vibrations applied to the wall 93 by the second vibrator 34C reach the first sensor 34A via the floor 91. As shown in the third path PH3, vibrations applied to the wall 93 by the second vibrator 34C propagate, for example, downward (= negative Z-axis direction), propagate through the floor 91 to the left rear, and propagate through the wall 94 upward (= positive Z-axis direction) to reach the first sensor 34A. The dashed line, representing the fourth path PH4, shows an example of a path through which vibrations applied to the wall 94 by the second vibrator 34C reach the second sensor 34B via the floor 91. As shown in the fourth path PH4, the vibration applied to the wall 93 by the second vibrator 34C propagates, for example, downward (= negative Z-axis direction), propagates through the floor 91 to the left front, and propagates through the wall 94 upward (= positive Z-axis direction) to reach the second sensor 34B.

[0039] The first sensor 34A outputs the detected vibration signal VB1 to the control device 4. When at least one of the first vibrator 24 and the second vibrator 34C applies vibration at a first frequency FQ1, the first sensor 34A outputs the detected vibration signal VB11 to the control device 4. When at least one of the first vibrator 24 and the second vibrator 34C applies vibration at a second frequency FQ2, the first sensor 34A outputs the detected vibration signal VB12 to the control device 4.

[0040] The second sensor 34B outputs the detected vibration signal VB2 to the control device 4. When at least one of the first vibrator 24 and the second vibrator 34C applies vibration at the first frequency FQ1, the second sensor 34B outputs the detected vibration signal VB21 to the control device 4. When at least one of the first vibrator 24 and the second vibrator 34C applies vibration at the second frequency FQ2, the second sensor 34B outputs the detected vibration signal VB22 to the control device 4. The first frequency FQ1 and the second frequency FQ2 will be further explained with reference to Figure 1.

[0041] In the first embodiment, a case where the first vibrator 24 is disposed on the switch 20 and the second vibrator 34C is disposed on the outlet 30C will be described, but the embodiment is not limited thereto. The first vibrator 24 and the second vibrator 34C may only need to be disposed on "electrical installation materials" fixed to a wall surface.

[0042] Further, in the first embodiment, a case where the first sensor 34A is disposed on the outlet 30A and the second sensor 34B is disposed on the outlet 30B will be described, but the embodiment is not limited thereto. The first sensor 34A and the second sensor 34B may only need to be disposed on "electrical installation materials" fixed to a wall surface.

[0043] [1-1-3. Configuration of Each Device] Next, the configuration of each device of the object detection system 1 will be described with reference to FIG. 3. FIG. 3 is a block diagram illustrating an example of the configuration of each device of the object detection system 1, and shows the configurations of the lighting device 10, the switch 20, the outlets 30A, 30B, 30C, and the distribution board 50. FIG. 3 is a diagram schematically showing the configuration of each device, and does not limit the detailed circuit configuration. For example, each device in FIG. 3 may include circuits not shown in FIG. 3.

[0044] The object detection system 1 includes a control device 4 that estimates at least one of the presence or absence of an object placed on the floor, the position of the object, and the type of the object, based on the detection results of the first sensor 34A and the second sensor 34B. The switch 20 transmits the detection result of a vibration sensor (not shown) that detects the vibration applied to the wall 94 by the first vibrator 24 to the control device 4. The outlet 30A transmits the detection result of the vibration sensor 34A to the control device 4. The outlet 30B transmits the detection result of the vibration sensor 34B to the control device 4. The outlet 30C transmits the detection result of a vibration sensor (not shown) that detects the vibration applied to the wall 93 by the second vibrator 34C to the control device 4.

[0045] The control device 4 can be provided in any one of the lighting device 10, the switch 20, the outlets 30A, 30B, 30C, and the distribution board 50. Alternatively, the control device 4 may be implemented in a device separate from these devices. In the first embodiment, a configuration in which the lighting device 10 includes the control device 4 will be described as an example.

[0046] A service line 58 is connected to the distribution board 50. The service line 58 is a power line led in from outside the building including the living space 9, or a main wiring of the building including the living space 9, and supplies commercial AC power to the distribution board 50.

[0047] As shown in FIG. 3, the distribution board 50 includes a main breaker 51, branch wirings 52, and circuit breakers 53A, 53B, 53C, 53D. The main breaker 51 is connected to the service line 58. The branch wirings 52 connect the main breaker 51 to each of the circuit breakers 53A, 53B, 53C, 53D. Embodiment 1 shows an example in which a single-phase three-wire 200V AC power is supplied to the distribution board 50 from the service line 58. In this example, the branch wirings 52 include branch wirings 52R, 52N, 52T respectively corresponding to the R phase, N phase, and T phase. The branch wirings 52R, 52N, 52T are configured by stranded wires, single-core wires, or metal members such as bus bars. In addition, the distribution board 50 may be provided with a ground terminal (not shown).

[0048] The main breaker 51 disconnects the service line 58 and the branch wirings 52 when the current flowing from the service line 58 to the branch wirings 52 exceeds a specified capacity. The distribution board 50 may include, in addition to the main breaker 51, an earth leakage breaker (not shown), a neutral wire open-phase protection circuit, and the like.

[0049] Circuit breaker 53A is connected to branch wiring 52R and branch wiring 52N and supplies single-phase 100V AC current to power line 5A. Circuit breaker 53A disconnects power line 5A from branch wiring 52 when the current flowing through power line 5A exceeds a predetermined capacity. Circuit breaker 53B is connected to branch wiring 52N and branch wiring 52T and supplies single-phase 100V AC current to power line 5B. Circuit breaker 53B disconnects power line 5B from branch wiring 52 when the current flowing through power line 5B exceeds a predetermined capacity. Circuit breaker 53C is connected to branch wiring 52N and branch wiring 52T and supplies single-phase 100V AC current to power line 5C. Circuit breaker 53C disconnects power line 5C from branch wiring 52 when the current flowing through power line 5C exceeds a predetermined capacity. The circuit breaker 53D is connected to branch wiring 52N and branch wiring 52T and supplies single-phase 100V AC current to power line 5D. The circuit breaker 53D disconnects power line 5D from branch wiring 52 when the current flowing through power line 5D exceeds a predetermined capacity.

[0050] The branch wiring 52 can be called secondary wiring with respect to the main breaker 51. That is, the main breaker 51 functions as a circuit breaker that separates the commercial power supply from the secondary wiring. Also, the power line 5 can be called secondary wiring with respect to the circuit breakers 53A, 53B, 53C, and 53D, and the circuit breakers 53A, 53B, 53C, and 53D function as circuit breakers that separate the commercial power supply from the secondary wiring.

[0051] In the object detection system 1, the lighting device 10 and the switch 20 are connected to power line 5A, the outlet 30A is connected to power line 5B, the outlet 30B is connected to power line 5C, and the outlet 30C is connected to power line 5D. Power lines 5A, 5B, 5C, and 5D are all connected to the branch wiring 52 of the distribution board 50. In other words, the lighting device 10, the switch 20, and the outlets 30A, 30B, and 30C are electrically connected to each other via the branch wiring 52 and power line 5.

[0052] In the object detection system 1, both wired and wireless communication methods can be used for communication between the lighting device 10, the switch 20, and the outlets 30A, 30B, and 30C. For wired communication, power line communication (PLC) via the power line 5 can be used. For wireless communication, Wi-Fi®, Bluetooth®, or other wireless communication methods can be used. Embodiment 1 describes an example in which power line communication via the power line 5 is used in the object detection system 1.

[0053] The lighting device 10 comprises a light source 11 and a driver 12 that lights the light source 11. The light emitted from the light source 11 is diffused by the cover 110 shown in Figure 2, illuminating the living space 9. The light source 11 is a solid light source such as an LED (Light Emitting Diode), or a lamp such as an incandescent lamp or a fluorescent lamp. The driver 12 is connected to the light source 11, and power is supplied from the driver 12 to the light source 11.

[0054] The driver 12 is connected to power line 5A and receives commercial AC power through power line 5A. The driver 12 includes a circuit that supplies the power required for the light source 11 to light up. For example, if the light source 11 is an LED, the driver 12 includes a converter circuit that converts the commercial AC power to output a DC current of a predetermined voltage to the light source 11. The driver 12 may also have a function to adjust the light intensity of the light source 11. For example, the driver 12 may be configured to adjust the light intensity of the light source 11 by PWM (Pulse Width Modulation) control.

[0055] The communication unit 16 and the power supply unit 17 are incorporated as a control unit 13 into the circuit body 111 shown in Figure 2 of the lighting device 10. The power supply unit 17 is connected to the power line 5A via a ceiling socket 112, and converts the AC current supplied from the power line 5A into a DC current of a predetermined voltage and outputs it to the communication unit 16.

[0056] The control unit 13 is equipped with a control device 4. The control device 4 includes a processor 41 and a memory 42.

[0057] The processor 41 processes data by executing a program. The processor 41 is composed of integrated circuits such as a CPU (Central Processing Unit), an MPU (Micro Processing Unit), and a microcontroller. The processor 41 may also be programmed hardware. The memory 42 includes a storage device that non-volatilely stores programs and data using flash memory, magnetic recording media, optical recording media, etc. The memory 42 may be composed of a SoC (System on a Chip) integrated with the processor 41. The processing performed by the processor 41 will be described later with reference to Figure 1.

[0058] The communication unit 16 is a communication device that performs power line communication with other devices constituting the object detection system 1 in accordance with instructions from the control device 4, and includes a transmitter, receiver, interface circuit, etc.

[0059] The communication unit 16 is, for example, a communication modem connected to two cables constituting the power line 5A, and performs power line communication via the power line 5A. The communication unit 16 includes, for example, an encoding circuit for encoding data, a transmitting circuit for superimposing the encoded signal onto a carrier wave, a receiving circuit equipped with a filter for extracting the signal superimposed on the carrier wave, and a decoding circuit for decoding the extracted signal to generate received data and outputting the received data to the control unit 15. Alternatively, the communication unit 16 may be a wireless communication device that performs wireless communication. The communication unit 16 receives detection data transmitted by the switch 20 and outlets 30A, 30B, and 30C. The detection data received by the communication unit 16 is output to the control device 4.

[0060] The switch 20 comprises a switch body 21 and a sensor unit 23. Figure 3 shows a single-pole switch as the switch body 21, which disconnects only one of the pair of cables constituting the power line 5A, but the switch body 21 may also be a double-pole switch.

[0061] The sensor unit 23 comprises a first vibrator 24, a control unit 25, a communication unit 26, and a power supply unit 27. The power supply unit 27 is connected to the power line 5A at a position closer to the control device 4 than the switch body 21. The power supply unit 27 converts the alternating current supplied from the power line 5A into a direct current of a predetermined voltage and outputs it to the first vibrator 24, the control unit 25, and the communication unit 26.

[0062] The control unit 25 includes a processor that processes data by executing a program. The processor is composed of an integrated circuit such as a CPU, MPU, or microcontroller. The control unit 25 may also include a storage device that non-volatilely stores programs and data using flash memory, magnetic recording media, optical recording media, etc. The control unit 25 may be composed of an integrated circuit that integrates the processor and the storage device, or it may be composed of an SoC that further integrates the first vibrator 24.

[0063] The control unit 25 operates the first vibrator 24 to apply vibration to the wall 94. The control unit 25 also acquires the detection result from a vibration sensor (not shown) located on the first vibrator 24 and generates detection data including the detection result. The control unit 25 transmits the detection data to the control device 4 via the communication unit 26.

[0064] The communication unit 26 communicates with the control device 4 according to instructions from the control unit 25. For example, the communication unit 26 transmits detection data generated by the control unit 25 to the control device 4. The communication unit 26 is a communication modem connected to two cables that make up the power line 5A and performs power line communication via the power line 5A. The communication unit 26 includes, for example, an encoding circuit for encoding data, a transmitting circuit for superimposing the encoded signal onto a carrier wave, a receiving circuit equipped with a filter for extracting the signal superimposed on the carrier wave, and a decoding circuit for decoding the extracted signal to generate received data and outputting the received data to the control unit 25.

[0065] The outlet 30A comprises an outlet body 31A and a sensor unit 33A. The outlet body 31A has a pair of terminals 32A that are connected to the power line 5B. Load equipment is connected to terminals 32A. Terminals 32A correspond to an example of power terminals. Load equipment is, for example, an electrical appliance used in the living space 9.

[0066] The sensor unit 33A comprises a first sensor 34A, a control unit 35A, a communication unit 36A, and a power supply unit 37A. The power supply unit 37A is connected to the power line 5B and converts the alternating current supplied from the power line 5B into a direct current of a predetermined voltage, which is then output to the first sensor 34A, the control unit 35A, and the communication unit 36A. For convenience, in Figure 3, the first sensor 34A is referred to simply as sensor 34A.

[0067] The control unit 35A includes a processor that processes data by executing a program. The processor is composed of an integrated circuit such as a CPU, MPU, or microcontroller. The control unit 35A may also include a storage device that non-volatilely stores programs and data using flash memory, magnetic recording media, optical recording media, etc. The control unit 35A may be composed of an integrated circuit that integrates the processor and the storage device, or it may be composed of an SoC that further integrates the first sensor 34A.

[0068] The control unit 35A operates the first sensor 34A to detect vibrations in the wall 94. The control unit 35A acquires the detection result from the first sensor 34A and generates detection data including the detection result. The control unit 35A transmits the detection data to the control device 4 via the communication unit 36A.

[0069] The communication unit 36A communicates with the control device 4 according to instructions from the control unit 35A. For example, the communication unit 36A transmits detection data generated by the control unit 35A to the control device 4. The communication unit 36A is a communication modem connected to two cables that constitute the power line 5B and performs power line communication via the power line 5B. The communication unit 36A includes, for example, an encoding circuit for encoding data, a transmitting circuit for superimposing the encoded signal onto a carrier wave, a receiving circuit equipped with a filter for extracting the signal superimposed on the carrier wave, and a decoding circuit for decoding the extracted signal to generate received data and outputting the received data to the control unit 35A.

[0070] Outlet 30B can have the same configuration as outlet 30A. That is, outlet 30B comprises an outlet body 31B and a sensor unit 33B, and the outlet body 31B has a pair of terminals 32B connected to the power line 5B. Terminals 32B correspond to an example of power terminals. Load equipment is connected to terminals 32B.

[0071] The sensor unit 33B includes a second sensor 34B, a control unit 35B, a communication unit 36B, and a power supply unit 37B. The power supply unit 37B is connected to the power line 5C and converts the alternating current supplied from the power line 5C into a direct current of a predetermined voltage, which it outputs to the second sensor 34B, the control unit 35B, and the communication unit 36B.

[0072] The control unit 35B includes a processor that processes data by executing a program. The processor is composed of an integrated circuit such as a CPU, MPU, or microcontroller. The control unit 35B may also include a storage device that non-volatilely stores programs and data using flash memory, magnetic recording media, optical recording media, etc. The control unit 35B may be composed of an integrated circuit that integrates the processor and the storage device, or it may be composed of an SoC that further integrates the second sensor 34B.

[0073] The control unit 35B activates the second sensor 34B to detect vibrations in the wall 95. The control unit 35B acquires the detection result from the second sensor 34B and generates detection data including the detection result. The control unit 35B transmits the detection data to the control device 4 via the communication unit 36B.

[0074] The communication unit 36B communicates with the control device 4 according to instructions from the control unit 35B. For example, the communication unit 36B transmits detection data generated by the control unit 35B to the control device 4. The communication unit 36B is a communication modem connected to two cables that constitute the power line 5B and performs power line communication via the power line 5B. The communication unit 36B includes, for example, an encoding circuit for encoding data, a transmitting circuit for superimposing the encoded signal onto a carrier wave, a receiving circuit equipped with a filter for extracting the signal superimposed on the carrier wave, and a decoding circuit for decoding the extracted signal to generate received data and outputting the received data to the control unit 35B.

[0075] The outlet 30C comprises an outlet body 31C and a sensor unit 33C. Figure 3 shows a single-pole switch as the outlet body 31C, which disconnects only one of the pair of cables that make up the power line 5, but the outlet body 31C may also be a double-pole switch.

[0076] The sensor unit 23 includes a second vibrator 34C, a control unit 35C, a communication unit 36C, and a power supply unit 37C. The power supply unit 37C is connected to the power line 5D at a position closer to the control device 4 than the outlet body 31C. The power supply unit 37C converts the alternating current supplied from the power line 5A into a direct current of a predetermined voltage and outputs it to the second vibrator 34C, the control unit 35C, and the communication unit 36C.

[0077] The control unit 35C includes a processor that processes data by executing a program. The processor is composed of an integrated circuit such as a CPU, MPU, or microcontroller. The control unit 35C may also include a storage device that non-volatilely stores programs and data using flash memory, magnetic recording media, optical recording media, etc. The control unit 35C may be composed of an integrated circuit that integrates the processor and the storage device, or it may be composed of an SoC that further integrates the second vibrator 34C.

[0078] The control unit 35C operates the second vibrator 34C to apply vibration to the wall 93. The control unit 35C also acquires the detection result from a vibration sensor (not shown) located on the second vibrator 34C and generates detection data including the detection result. The control unit 35C causes the control device 4 to transmit the detection data via the communication unit 36C.

[0079] The communication unit 36C communicates with the control device 4 according to instructions from the control unit 25. For example, the communication unit 36C transmits detection data generated by the control unit 25 to the control device 4. The communication unit 36C is a communication modem connected to two cables that constitute the power line 5D and performs power line communication via the power line 5D. The communication unit 36C includes, for example, an encoding circuit for encoding data, a transmitting circuit for superimposing the encoded signal onto a carrier wave, a receiving circuit equipped with a filter for extracting the signal superimposed on the carrier wave, and a decoding circuit for decoding the extracted signal to generate received data and outputting the received data to the control unit 25.

[0080] The control device 4 installed in the lighting device 10 operates by receiving power from the power supply unit 17. The processor 41 also directly or indirectly controls the communication unit 16 when communicating with the switch 20 and the outlets 30A, 30B, and 30C. The processor 41 obtains detection data including the detection result of a vibration sensor (not shown) located in the first vibrator 24, the detection result of a vibration sensor (not shown) located in the second vibrator 34, the detection data including the detection result of the first sensor 34A, and the detection data including the detection result of the second sensor 34B, via the communication unit 16.

[0081] [1-1-4. Configuration of the Control Device] Next, the configuration of the control device 4 will be described with reference to Figure 1. As described with reference to Figure 3, the control device 4 is configured to communicate with, for example, the vibrator 6 and the sensor 7 via power line communication. The vibrator 6 includes a first vibrator 24 and a second vibrator 34C. The first vibrator 24 is located on the switch 20. The second vibrator 34C is located on the outlet 30C. The sensor 7 includes a first sensor 34A and a second sensor 34B. The first sensor 34A is located on the outlet 30A. The second sensor 34B is located on the outlet 30B. Based on the detection results of the first sensor 34A and the second sensor 34B, the control device 4 estimates the position of an object placed on the floor. The control device 4 also notifies the smartphone 8 of the estimation result. The "object" is, for example, a desk DK and a human body P.

[0082] The control device 4 comprises a processor 41 and a memory 42. The processor 41 is a processor such as a CPU or MPU. The memory 42 is a memory for storing programs and data. The memory 42 stores the control program 421.

[0083] The memory 42 has a non-volatile storage area. Alternatively, the memory 42 may also have a volatile storage area and constitute the work area of ​​the processor 41. The memory 42 is composed of, for example, ROM or RAM.

[0084] The processor 41 functions as a setting unit 411, an acquisition unit 412, an estimation unit 413, and a notification unit 414 by reading and executing the control program 421. The processor 41 also causes the memory 42 to function as a first model storage unit 422, a second model storage unit 423, and a data storage unit 424 by reading and executing the control program 421.

[0085] The data storage unit 425 stores multiple sets of input datasets DJ1 and multiple sets of output datasets DQ1 used to generate a first trained model M1 by performing supervised learning using deep learning. The data storage unit 425 also stores multiple sets of input datasets DJ2 and multiple sets of output datasets DQ2 used to generate a second trained model M2 by performing supervised learning using deep learning. The first trained model M1 and the second trained model M2 correspond to examples of "trained models".

[0086] Each of the multiple input datasets DJ1 used to generate the first trained model M1 includes, for example, an applied vibration signal VA11, an applied vibration signal VA21, a detected vibration signal VB11, and a detected vibration signal VB21. The applied vibration signal VA11 is a vibration signal detected by a vibration sensor (not shown) located on the first vibrator 24 when the first vibrator 24 applies vibration at a first frequency FQ1. The applied vibration signal VA21 is a vibration signal detected by a vibration sensor (not shown) located on the second vibrator 34C when the second vibrator 34C applies vibration at a first frequency FQ1. The detected vibration signal VB11 is a vibration signal detected by the first sensor 34A when at least one of the first vibrator 24 and the second vibrator 34C applies vibration at a first frequency FQ1. The detected vibration signal VB21 is a vibration signal detected by the second sensor 34B when at least one of the first vibrator 24 and the second vibrator 34C applies vibration of the first frequency FQ1.

[0087] Each of the multiple sets of output datasets DQ1 used to generate the first trained model M1 includes, for example, position XD and position YD. Position XD is the X-coordinate of the position of desk DK, and position YD is the Y-coordinate of the position of desk DK. The first trained model M1 is generated in advance by, for example, the control device 4 and stored in the first model storage unit 422. The first trained model M1 is read out by the estimation unit 413 and executed. The first trained model M1 will be further explained with reference to Figure 4.

[0088] Multiple sets can be, for example, 10,000 sets. For example, there are 10 types of applied vibration signals VA11, 10 types of applied vibration signals VA21, 10 types of position XD, and 10 types of position YD (10,000 = 10 × 10 × 10 × 10). The input dataset DJ1 and output dataset DQ1 are determined experimentally.

[0089] Each of the multiple input datasets DJ2 used to generate the second trained model M2 includes, for example, an applied vibration signal VA12, an applied vibration signal VA22, a detected vibration signal VB12, and a detected vibration signal VB22. The applied vibration signal VA12 is a vibration signal detected by a vibration sensor (not shown) located on the first vibrator 24 when the first vibrator 24 applies vibration at a second frequency FQ2. The applied vibration signal VA22 is a vibration signal detected by a vibration sensor (not shown) located on the second vibrator 34C when the second vibrator 34C applies vibration at a second frequency FQ2. The detected vibration signal VB12 is a vibration signal detected by the first sensor 34A when at least one of the first vibrator 24 and the second vibrator 34C applies vibration at a second frequency FQ2. The detected vibration signal VB22 is a vibration signal detected by the second sensor 34B when at least one of the first vibrator 24 and the second vibrator 34C applies vibration of the second frequency FQ2.

[0090] Each of the multiple sets of output datasets DQ2 used to generate the second trained model M2 includes, for example, position XP and position YP. Position XP is the X-coordinate of the position of the human body P, and position YD is the Y-coordinate of the position of the human body P. The second trained model M2 is generated in advance by, for example, the control device 4 and stored in the second model storage unit 423. The second trained model M2 is read out by the estimation unit 413 and executed. The second trained model M2 will be further explained with reference to Figure 5.

[0091] Multiple sets can be, for example, 1,000,000 sets. For example, there are 10 types of applied vibration signals VA12, 10 types of applied vibration signals VA22, 10 types of position XP, and 10 types of position YP (10,000 = 10 × 100 × 10 × 10). The input dataset DJ2 and output dataset DQ2 are determined experimentally.

[0092] The first model storage unit 422 pre-stores the first trained model M1. The first trained model M1 is read out by the estimation unit 413 and executed. The first model storage unit 422 corresponds to an example of a "model storage unit".

[0093] The second model storage unit 423 stores the second trained model M2 in advance. The second trained model M2 is read out by the estimation unit 413 and executed. The second model storage unit 423 corresponds to an example of a "model storage unit".

[0094] The setting unit 411 sets the frequency FQ of the vibration applied by the vibrator 6 according to the object whose position the estimation unit 413 is estimating. The object whose position the estimation unit 413 is estimating is set, for example, based on user input. For example, if the object whose position the estimation unit 413 is estimating is a desk DK, the frequency FQ of the vibration applied by the vibrator 6 is set to a first frequency FQ1. The first frequency FQ1 is, for example, the resonant frequency of the desk DK, or a harmonic frequency of the resonant frequency of the desk DK. For example, if the object whose position the estimation unit 413 is estimating is a human body P, the frequency FQ of the vibration applied by the vibrator 6 is set to a second frequency FQ2. The second frequency FQ2 is, for example, the resonant frequency of the human body P, or a harmonic frequency of the resonant frequency of the human body P.

[0095] In Embodiment 1, the setting unit 411 was described as setting the frequency FQ of the vibration applied by the vibrator 6 according to the object whose position is estimated by the estimation unit 413, but the embodiments are not limited to this. The setting unit 411 may, for example, set the range of the vibration frequency applied by the vibrator 6 according to the object whose position is estimated by the estimation unit 413. In this case, the vibrator 6 changes the frequency of the vibration applied by the vibrator 6 from, for example, the lower limit to the upper limit of the frequency range.

[0096] The acquisition unit 412 acquires detection signals from the vibrator 6 and the sensor 7 according to the object whose position the estimation unit 413 is estimating. When the object whose position the estimation unit 413 is estimating is a desk DK, the acquisition unit 412 acquires the following four signals. That is, the acquisition unit 412 acquires the applied vibration signal VA11 from the first vibrator 24, the applied vibration signal VA21 from the second vibrator 34C, the detected vibration signal VB11 from the first sensor 34A, and the detected vibration signal VB21 from the second sensor 34B. When the object whose position the estimation unit 413 is estimating is a human body P, the acquisition unit 412 acquires the following four signals. Specifically, the acquisition unit 412 acquires the applied vibration signal VA12 from the first vibrator 24, the applied vibration signal VA22 from the second vibrator 34C, the detected vibration signal VB12 from the first sensor 34A, and the detected vibration signal VB22 from the second sensor 34B.

[0097] The estimation unit 413 estimates the position of an object placed on the floor 91 based on the detection results of the sensor 7. For example, the estimation unit 413 estimates the position of a desk DK placed on the floor 91, i.e., position XD and position YD, using the first trained model M1. The first trained model M1 is a trained model M that estimates the position of the desk DK. The estimation unit 413 reads the first trained model M1 stored in the first model storage unit 422 and inputs the applied vibration signal VA11, applied vibration signal VA21, detected vibration signal VB11, and detected vibration signal VB21 acquired by the acquisition unit 412 to the first trained model M1. Then, the estimation unit 413 estimates the position XD and position YD output from the first trained model M1 as the position of the desk DK.

[0098] Furthermore, the estimation unit 413 estimates, for example, the position of the human body P on the floor 91, i.e., position XP and position YP, using the second trained model M2. The second trained model M2 is a trained model M that estimates the position of the human body P. The estimation unit 413 reads the second trained model M2 stored in the second model storage unit 423 and inputs the applied vibration signal VA12, applied vibration signal VA22, detected vibration signal VB12, and detected vibration signal VB22 acquired by the acquisition unit 412 to the second trained model M2. The estimation unit 413 then estimates the position XP and position YP output from the second trained model M2 as the position of the human body P.

[0099] The notification unit 414 notifies the estimation results of the estimation unit 413. If the estimation unit 413 estimates the position of desk DK, i.e., position XD and position YD, the notification unit 414 transmits position XD and position YD to the smartphone 8. If the estimation unit 413 estimates the position of human body P, i.e., position XP and position YP, the notification unit 414 transmits position XP and position YP to the smartphone 8.

[0100] [1-1-5. Configuration of the Trained Model] Next, the configuration of the trained model M will be described with reference to Figures 4 and 5. Figure 4 is a diagram showing an example of the first trained model M1 in Embodiment 1. As shown in Figure 4, the first trained model M1 is composed of a multi-layer artificial neural network. The first trained model M1 is a model that estimates the position of desk DK, that is, position XD and position YD.

[0101] The nodes constituting the input layer receive the applied vibration signal VA11, the applied vibration signal VA21, the detected vibration signal VB11, and the detected vibration signal VB21. The applied vibration signal VA11 is the vibration signal detected by a vibration sensor (not shown) located on the first vibrator 24 when the first vibrator 24 applies vibration at a first frequency FQ1. The applied vibration signal VA21 is the vibration signal detected by a vibration sensor (not shown) located on the second vibrator 34C when the second vibrator 34C applies vibration at a first frequency FQ1. The detected vibration signal VB11 is the vibration signal detected by the first sensor 34A when at least one of the first vibrator 24 and the second vibrator 34C applies vibration at a first frequency FQ1. The detected vibration signal VB21 is a vibration signal detected by the second sensor 34B when at least one of the first vibrator 24 and the second vibrator 34C applies vibration of the first frequency FQ1.

[0102] The nodes constituting the output layer output the position of desk DK, i.e., position XD and position YD. Position XD is the X-coordinate of desk DK's position, and position YD is the Y-coordinate of desk DK's position.

[0103] The first trained model M1 is generated by performing supervised learning using deep learning with multiple input datasets DJ1 and multiple output datasets DQ1. The number of sets can be, for example, 10,000.

[0104] Figure 5 shows an example of the second trained model M2 in Embodiment 1. As shown in Figure 5, the second trained model M2 is composed of a multi-layer artificial neural network. The second trained model M2 is a model that estimates the position of the human body P, i.e., position XP and position YP.

[0105] The nodes constituting the input layer receive the applied vibration signal VA12, the applied vibration signal VA22, the detected vibration signal VB12, and the detected vibration signal VB22. The applied vibration signal VA12 is the vibration signal detected by a vibration sensor (not shown) located on the first vibrator 24 when the first vibrator 24 applies vibration at the second frequency FQ2. The applied vibration signal VA22 is the vibration signal detected by a vibration sensor (not shown) located on the second vibrator 34C when the second vibrator 34C applies vibration at the second frequency FQ2. The detected vibration signal VB12 is the vibration signal detected by the second sensor 34B when at least one of the first vibrator 24 and the second vibrator 34C applies vibration at the second frequency FQ2. The detected vibration signal VB22 is a vibration signal detected by the second sensor 34B when at least one of the first vibrator 24 and the second vibrator 34C applies vibration of the second frequency FQ2.

[0106] The nodes constituting the output layer output the position of the human body P, i.e., position XP and position YP. Position XP is the X-coordinate of the human body P's position, and position YP is the Y-coordinate of the human body P's position.

[0107] The second trained model M2 is generated by performing supervised learning using deep learning with multiple input datasets DJ2 and multiple output datasets DQ2. The number of sets can be, for example, 10,000.

[0108] Embodiment 1 describes a case in which each of the first trained model M1 and the second trained model M2 is generated by supervised learning using deep learning, but the embodiments are not limited to this. Each of the first trained model M1 and the second trained model M2 may be generated by "machine learning" or the like. That is, each of the first trained model M1 and the second trained model M2 may be generated by, for example, "reinforcement learning" or "deep reinforcement learning".

[0109] [1-2. Processing of the Control Device] Next, the processing of the control device 4 will be explained with reference to Figure 6. Figure 6 is a flowchart showing an example of the processing of the control device 4. The first trained model M1 is pre-generated by the control device 4 by performing supervised learning using deep learning with the input dataset DJ1 and the output dataset DQ1, and is stored in the first model storage unit 422. The second trained model M2 is also pre-generated by the control device 4 by performing supervised learning using deep learning with the input dataset DJ2 and the output dataset DQ2, and is stored in the second model storage unit 423.

[0110] As shown in Figure 6, in step S101, the setting unit 411 determines whether the object whose position is to be estimated is a desk DK. If the setting unit 411 determines that the object whose position is to be estimated is a desk DK (step S101; YES), the process proceeds to step S121. If the setting unit 411 determines that the object whose position is to be estimated is not a desk DK (step S101; NO), the process proceeds to step S103. Then, in step S103, the setting unit 411 determines whether the object whose position is to be estimated is a human body P. If the setting unit 411 determines that the object whose position is to be estimated is not a human body P (step S103; NO), the process returns to step S101. If the setting unit 411 determines that the object whose position is to be estimated is a human body P (step S101; YES), the process proceeds to step S105.

[0111] Then, in step S105, the setting unit 411 sets the frequency FQ of the vibrations that the first vibrator 24 and the second vibrator 34C apply to the wall to the second frequency FQ2, and instructs the first vibrator 24 and the second vibrator 34C to apply vibrations of the second frequency FQ2 to the wall. Next, in step S107, the acquisition unit 412 acquires the applied vibration signal VA12 from the first vibrator 24 and the applied vibration signal VA22 from the second vibrator 34C. The applied vibration signal VA12 is the vibration signal detected by a vibration sensor (not shown) located on the first vibrator 24 when the first vibrator 24 applies vibrations of the second frequency FQ2. The applied vibration signal VA22 is the vibration signal detected by a vibration sensor (not shown) located on the second vibrator 34C when the second vibrator 34C applies vibrations of the second frequency FQ2.

[0112] Next, in step S109, the acquisition unit 412 acquires a detected vibration signal VB12 from the first sensor 34A and a detected vibration signal VB22 from the second sensor 34B. The detected vibration signal VB12 is a vibration signal detected by the second sensor 34B when the first vibrator 24 and the second vibrator 34C apply vibration of the second frequency FQ2. The detected vibration signal VB22 is a vibration signal detected by the second sensor 34B when the first vibrator 24 and the second vibrator 34C apply vibration of the second frequency FQ2. Next, in step S111, the acquisition unit 412 determines whether a predetermined period PD has elapsed since the start of acquiring the detected vibration signals VB12 and VB22. The predetermined period PD is, for example, the sum of the excitation period PD1 of the first vibrator 24 and the second vibrator 34C and the vibration propagation period PD2. The excitation period PD1 is, for example, the period during which vibrations equivalent to 10 cycles are applied. The propagation period PD2 is the period during which the vibrations applied by the first vibrator 24 and the second vibrator 34C reach the first sensor 34A and the second sensor 34B.

[0113] If the acquisition unit 412 determines that a predetermined period PD has not elapsed since the start of acquiring the detected vibration signals VB12 and VB22 (step S111; NO), the process returns to step S105. If the acquisition unit 412 determines that a predetermined period PD has elapsed since the start of acquiring the detected vibration signals VB12 and VB22 (step S111; YES), the process proceeds to step S113. In step S113, the setting unit 411 stops the vibrations applied to the wall by the first vibrator 24 and the second vibrator 34C.

[0114] Next, in step S115, the estimation unit 413 reads the second trained model M2 stored in the second model storage unit 423 and inputs the applied vibration signal VA12, applied vibration signal VA22, detected vibration signal VB12, and detected vibration signal VB22 acquired by the acquisition unit 412 to the second trained model M2. Next, in step S117, the estimation unit 413 estimates the position XP and position YP output from the second trained model M2 as the position of the human body P. Next, in step S119, the notification unit 414 transmits the position XP and position YP to the smartphone 8 and displays the position XP and position YP on the smartphone 8. After that, the process proceeds to step S137.

[0115] If the answer in step S101 is YES, then in step S121, the setting unit 411 sets the frequency FQ of the vibration applied to the wall by the first vibrator 24 and the second vibrator 34C to a first frequency FQ1, and instructs the first vibrator 24 and the second vibrator 34C to apply vibration of the first frequency FQ1 to the wall. Next, in step S123, the acquisition unit 412 acquires the applied vibration signal VA11 from the first vibrator 24 and the applied vibration signal VA21 from the second vibrator 34C. The applied vibration signal VA11 is the vibration signal detected by a vibration sensor (not shown) located on the first vibrator 24 when the first vibrator 24 applies vibration of the first frequency FQ1. The applied vibration signal VA21 is the vibration signal detected by a vibration sensor (not shown) located on the second vibrator 34C when the second vibrator 34C applies vibration of the first frequency FQ1.

[0116] Next, in step S125, the acquisition unit 412 acquires a detected vibration signal VB11 from the first sensor 34A and a detected vibration signal VB21 from the second sensor 34B. The detected vibration signal VB11 is a vibration signal detected by the second sensor 34B when the first vibrator 24 and the second vibrator 34C apply vibration at a first frequency FQ1. The detected vibration signal VB21 is a vibration signal detected by the second sensor 34B when the first vibrator 24 and the second vibrator 34C apply vibration at a first frequency FQ1. Next, in step S127, the acquisition unit 412 determines whether a predetermined period PD has elapsed since the start of acquiring the detected vibration signals VB11 and VB21. The predetermined period PD is, for example, the sum of the excitation period PD1 of the first vibrator 24 and the second vibrator 34C and the vibration propagation period PD2. The excitation period PD1 is, for example, the period during which vibrations equivalent to 10 cycles are applied. The propagation period PD2 is the period during which the vibrations applied by the first vibrator 24 and the second vibrator 34C reach the first sensor 34A and the second sensor 34B.

[0117] If the acquisition unit 412 determines that a predetermined period PD has not elapsed since the start of acquiring the detected vibration signals VB11 and VB21 (step S127; NO), the process returns to step S121. If the acquisition unit 412 determines that a predetermined period PD has elapsed since the start of acquiring the detected vibration signals VB11 and VB21 (step S127; YES), the process proceeds to step S129. In step S129, the setting unit 411 stops the vibrations applied to the wall by the first vibrator 24 and the second vibrator 34C.

[0118] Next, in step S131, the estimation unit 413 reads the first trained model M1 stored in the first model storage unit 422 and inputs the applied vibration signal VA11, applied vibration signal VA21, detected vibration signal VB11, and detected vibration signal VB21 acquired by the acquisition unit 412 to the first trained model M1. Next, in step S133, the estimation unit 413 estimates the position XD and position YD output from the first trained model M1 as the position of the desk DK. Next, in step S135, the notification unit 414 transmits the position XD and position YD to the smartphone 8 and displays the position XD and position YD on the smartphone 8. After that, the process proceeds to step S137.

[0119] Then, in step S137, the control device 4 determines whether or not to terminate the position detection process, which is the process of detecting the position of an object. The control device 4 determines whether or not to terminate the position detection process, which is the process of detecting the position of an object, based on, for example, an operation from the user. If the control device 4 determines that the position detection process will not be terminated (step S137; NO), the process returns to step S101. If the control device 4 determines that the position detection process will be terminated (step S137; YES), the process then ends.

[0120] [1-3. Effects, etc.] As described above, the object detection system 1 of Embodiment 1 includes a first vibrator 24 and a second vibrator 34C that apply vibration to at least one of the floor 91 and the walls 93-96, a first sensor 34A and a second sensor 34B that detect vibrations propagating through at least one of the floor 91 and the walls 93-96, and a control device 4 that controls the first vibrator 24 and the second vibrator 34C. The control device 4 includes an estimation unit 413 that estimates at least one of the presence or absence of an object placed on the floor 91, the position of the object, and the type of the object, based on the detection results of the first sensor 34A and the second sensor 34B, and a notification unit 414 that notifies the estimation results of the estimation unit 413.

[0121] In this configuration, the first vibrator 24 and the second vibrator 34C apply vibration to at least one of the floor 91 and the walls 93-96, and the first sensor 34A and the second sensor 34B detect the vibration propagating through at least one of the floor 91 and the walls 93-96. Based on the detection results of the first sensor 34A and the second sensor 34B, at least one of the presence or absence of an object placed on the floor 91, the position of the object, and the type of the object is estimated. Therefore, with a simple configuration, at least one of the presence or absence of an object placed on the floor 91, the position of the object, and the type of object can be accurately estimated. In addition, the user convenience can be improved by notifying the user of the estimation results.

[0122] In the object detection system 1 described above, the control device 4 uses the detection results of the first sensor 34A and the second sensor 34B as input datasets DJ1 and DJ2, and at least one of the presence or absence of an object, the position of the object, and the type of the object as output datasets DQ1 and DQ2. The control device 4 is equipped with model storage units 422 and 423 that generate a trained model M learned by machine learning and store the generated trained model M. The estimation unit 413 uses the trained model M to estimate at least one of the presence or absence of an object, the position of the object, and the type of the object.

[0123] With this configuration, the detection results from the first sensor 34A and the second sensor 34B are used as input datasets DJ1 and DJ2, and at least one of the presence or absence of an object, the position of the object, and the type of the object is used as output datasets DQ1 and DQ2. A trained model M learned by machine learning is generated, thus enabling the generation of an appropriate trained model M. Furthermore, by using the appropriate trained model M, at least one of the presence or absence of an object, the position of the object, and the type of object can be estimated appropriately.

[0124] In the object detection system 1 described above, the control device 4 includes a setting unit 411 that sets the frequency FQ of the vibration to be applied to each of the first vibrator 24 and the second vibrator 34C according to the type of object.

[0125] With this configuration, the frequency FQ of the vibration to be applied is set for each of the first vibrator 24 and the second vibrator 34C according to the type of object. By appropriately setting the frequency FQ of the vibration to be applied, it is possible to appropriately estimate at least one of the presence or absence of an object, the position of the object, and the type of object.

[0126] In the object detection system 1 described above, the first vibrator 24, the second vibrator 34C, the first sensor 34A, the second sensor 34B, and the control device 4 are each connected to a commercial power supply and are placed on electrical equipment that is permanently installed in the living space.

[0127] In this configuration, the first vibrator 24, the second vibrator 34C, the first sensor 34A, the second sensor 34B, and the control device 4 are each connected to a commercial power supply and placed in electrical equipment that is permanently installed in the living space. Therefore, power can be supplied to each of the first vibrator 24, the second vibrator 34C, the first sensor 34A, the second sensor 34B, and the control device 4 with a simple configuration. Furthermore, the first vibrator 24, the second vibrator 34C, the first sensor 34A, and the second sensor 34B can be configured to communicate with the control device 4 with a simple configuration.

[0128] In the object detection system 1 described above, the first vibrator 24 and the second vibrator 34C include vibration motors.

[0129] With this configuration, since the first vibrator 24 and the second vibrator 34C include vibration motors, the amplitude of the vibration applied to at least one of the floor 91 and the walls 93-96 can be increased. Therefore, the detection accuracy of the first sensor 34A and the second sensor 34B can be improved. Consequently, at least one of the presence or absence of an object, the position of the object, and the type of object can be accurately estimated.

[0130] In the object detection system 1 described above, the first vibrator 24 and the second vibrator 34C include piezoelectric elements.

[0131] With this configuration, since the first vibrator 24 and the second vibrator 34C include piezoelectric elements, vibration can be applied to at least one of the floor 91 and the walls 93-96 with a simple configuration. Therefore, the object detection system 1 can be configured with a simple configuration.

[0132] In the object detection system 1 described above, the first vibrator 24 and the second vibrator 34C include electrostrictive oscillators.

[0133] With this configuration, since the first vibrator 24 and the second vibrator 34C include electrostrictive oscillators, vibration can be applied to at least one of the floor 91 and the walls 93-96 with a simple configuration. Therefore, the object detection system 1 can be configured with a simple configuration.

[0134] In the object detection system 1 described above, the first vibrator 24 and the second vibrator 34C include magnetostrictive oscillators.

[0135] With this configuration, since the first vibrator 24 and the second vibrator 34C include magnetostrictive oscillators, vibration can be applied to at least one of the floor 91 and the walls 93-96 with a simple configuration. Therefore, the object detection system 1 can be configured with a simple configuration.

[0136] In the object detection system 1 described above, the first sensor 34A and the second sensor 34B include vibration sensors.

[0137] With this configuration, since the first sensor 34A and the second sensor 34B include vibration sensors, vibrations propagating through at least one of the floor 91 and the walls 93-96 can be properly detected with a simple configuration.

[0138] In the object detection system 1 described above, the first sensor 34A and the second sensor 34B include microphones.

[0139] With this configuration, since the first sensor 34A and the second sensor 34B include microphones, vibrations propagating through at least one of the floor 91 and the walls 93-96 can be properly detected with a simple configuration.

[0140] (Other Embodiments) As described above, the above embodiments have been explained as examples disclosed in this application. However, the technology in this disclosure is not limited to these embodiments and can be applied to embodiments that have been modified, replaced, added, or omitted. Furthermore, it is possible to create new embodiments by combining the components described in the above embodiments. Therefore, other embodiments are described below as examples.

[0141] In the above embodiment, the switch 20 and outlets 30A, 30B, and 30C are examples of "electrical equipment." The "electrical equipment" that constitutes the object detection system 1 can be, for example, outlets with two or more sockets, outlets with ground terminals, switches with pilot lamps, switches with timers, switches with lighting, ventilation fans, etc. It is also possible to incorporate a vibrator or vibration sensor into other "electrical equipment."

[0142] In the above embodiment, the control device 4 will be described in which it estimates the position of an object placed on the floor 91 based on the detection results of the first sensor 34A and the second sensor 34B, but the embodiment is not limited thereto. The control device 4 only needs to estimate at least one of the presence or absence of an object placed on the floor 91, the position of the object, and the type of the object, based on the detection results of the first sensor 34A and the second sensor 34B. For example, the control device 4 may detect the presence or absence of an object based on the detection results of the first sensor 34A and the second sensor 34B. Alternatively, the control device 4 may detect the type of object based on the detection results of the first sensor 34A and the second sensor 34B.

[0143] In the above embodiment, the case in which the first vibrator 24 and the second vibrator 34C apply vibration to the wall has been described, but the embodiment is not limited thereto. The first vibrator 24 and the second vibrator 34C may apply vibration to at least one of the floor 91 and the wall. When the first vibrator 24 and the second vibrator 34C apply vibration to the floor 91, reflection and attenuation of vibration between the wall and the floor 91 can be suppressed. Therefore, it is preferable for the first vibrator 24 and the second vibrator 34C to apply vibration to the floor 91.

[0144] In the above embodiment, the case in which each of the first sensor 34A and the second sensor 34B detects vibrations of the wall has been described, but the embodiment is not limited thereto. Each of the first sensor 34A and the second sensor 34B may detect vibrations propagating through at least one of the floor 91 and the wall. When each of the first sensor 34A and the second sensor 34B detects vibrations propagating through the floor 91, it is possible to suppress the reflection and attenuation of vibrations between the wall and the floor 91. Therefore, it is preferable that each of the first sensor 34A and the second sensor 34B detects vibrations propagating through the floor 91.

[0145] The processor 41 may consist of a single processor or multiple processors. These processors may also be hardware programmed to implement the corresponding functional units. That is, these processors may consist of, for example, an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0146] The configuration of each part of the object detection system 1 shown in Figures 1-3 is merely an example, and the specific implementation is not particularly limited. In other words, it is not necessarily required that hardware corresponding to each part in Figures 1 and 3 be implemented individually; for example, it is certainly possible to configure a system in which a single integrated circuit realizes the functions of each part. Furthermore, in the above-described embodiment, the functions of the processor 41 may be realized by software or by hardware.

[0147] The configuration of each part of the control device 4 shown in Figure 1 is merely an example, and the specific implementation is not particularly limited. In other words, it is not necessarily required that hardware corresponding to each part in Figure 1 be implemented individually; for example, it is certainly possible to configure a single integrated circuit to realize the functions of each part. Furthermore, in the above-described embodiment, the functions of the processor 41 may be realized by software or by hardware.

[0148] The processing steps shown in Figure 6 are divided according to the main processing content to facilitate understanding of the operation, and the operation is not limited by the way the processing units are divided or their names. Depending on the processing content, it may be further divided into more steps. Alternatively, it may be divided so that one step unit contains even more processing. Furthermore, the order of the steps may be rearranged as appropriate, as long as it does not impede the intent of this disclosure.

[0149] Since the embodiments described above are for illustrative purposes of the technology described herein, various modifications, substitutions, additions, omissions, etc., can be made within the claims or their equivalents.

[0150] (Note) The above description of embodiments discloses the following technology.

[0151] (Technology 1) An object detection system comprising: a plurality of vibration application mechanisms for applying vibration to at least one of the floor and the wall; a plurality of vibration detection mechanisms for detecting vibrations propagating through at least one of the floor and the wall; and a control device for controlling the plurality of vibration application mechanisms, wherein the control device comprises an estimation unit for estimating at least one of the presence or absence of an object placed on the floor, the position of the object, and the type of the object, based on the detection results of the plurality of vibration detection mechanisms; and a notification unit for notifying the estimation results of the estimation unit.

[0152] According to this system, multiple vibration application mechanisms apply vibration to at least one of the floor and / or wall, and multiple vibration detection mechanisms detect the vibrations propagating through at least one of the floor and / or wall. Based on the detection results from the multiple vibration detection mechanisms, the system estimates at least one of the following: the presence or absence of an object on the floor, the object's position, and the type of the object. Therefore, with a simple configuration, the system can accurately estimate the presence or absence of an object on the floor, its position, and at least one of the type of the object. Furthermore, by providing notification of the estimation results, user convenience can be improved.

[0153] (Technology 2) The object detection system according to Technology 1, wherein the control device includes a model storage unit that generates a trained model learned by machine learning using the detection results of the plurality of vibration detection mechanisms as an input dataset and at least one of the presence or absence of the object, the position of the object, and the type of the object as an output dataset, and stores the generated trained model, and the estimation unit estimates at least one of the presence or absence of the object, the position of the object, and the type of the object using the trained model.

[0154] According to this method, a properly trained model can be generated by using the detection results of multiple vibration detection mechanisms as the input dataset and at least one of the following as the output dataset: presence or absence of an object, position of the object, and type of the object. Furthermore, by using this properly trained model to estimate at least one of the following: presence or absence of an object, position of the object, and type of the object, it is possible to properly estimate at least one of these.

[0155] (Technology 3) The object detection system according to Technology 1 or Technology 2, wherein the control device includes a setting unit that sets the frequency of vibration to be applied to each of the plurality of vibration application mechanisms according to the type of object.

[0156] According to this, by setting the frequency of vibration to be applied to each of the multiple vibration application mechanisms according to the type of object, it is possible to appropriately estimate at least one of the presence or absence of an object, the position of the object, and the type of object by appropriately setting the frequency of vibration to be applied.

[0157] (Technology 4) The object detection system according to any one of Technology 1 to 3, wherein each of the plurality of vibration application mechanisms, the plurality of vibration detection mechanisms, and the control device is connected to a commercial power supply and is arranged on electrical equipment that is fixedly installed in a living space.

[0158] According to this design, each of the multiple vibration application mechanisms, multiple vibration detection mechanisms, and control devices is connected to a commercial power supply and positioned on electrical equipment permanently installed in the living space. Therefore, power can be supplied to each of the multiple vibration application mechanisms, multiple vibration detection mechanisms, and control devices with a simple configuration. Furthermore, each of the multiple vibration application mechanisms and multiple vibration detection mechanisms can be configured to communicate with the control device with a simple configuration.

[0159] (Technology 5) The object detection system according to any one of Technology 1 to Technology 4, wherein the vibration application mechanism includes a vibration motor.

[0160] According to this, since the vibration application mechanism includes a vibration motor, the amplitude of the vibration applied to at least one of the floor and the wall can be increased compared to the case where the vibration application mechanism includes, for example, a piezoelectric element. Therefore, the detection accuracy of multiple vibration detection mechanisms can be improved. Consequently, at least one of the presence or absence of an object, the position of the object, and the type of object can be accurately estimated.

[0161] (Technology 6) The object detection system according to any one of Technology 1 to Technology 4, wherein the vibration application mechanism includes a piezoelectric element.

[0162] According to this, since the vibration application mechanism includes a piezoelectric element, it can apply vibration to at least one of the floor and the wall with a simpler configuration compared to, for example, a vibration application mechanism that includes a vibration motor. Therefore, an object detection system can be configured with a simple configuration.

[0163] (Technology 7) The vibration application mechanism is an object detection system according to any one of Technology 1 to 4, wherein the vibration application mechanism includes an electrostrictive vibrator. With this, since the vibration application mechanism includes an electrostrictive vibrator, vibration can be applied to at least one of the floor and the wall with a simpler configuration compared to, for example, the case in which the vibration application mechanism includes a vibration motor. Therefore, an object detection system can be configured with a simple configuration.

[0164] (Technology 8) The vibration application mechanism is an object detection system according to any one of Technology 1 to 4, wherein the vibration application mechanism includes a magnetostrictive vibrator. With this, since the vibration application mechanism includes a magnetostrictive vibrator, vibration can be applied to at least one of the floor and the wall with a simpler configuration compared to the case where the vibration application mechanism includes, for example, a vibration motor. Therefore, an object detection system can be configured with a simple configuration.

[0165] (Technology 9) The vibration detection mechanism is an object detection system according to any one of Technology 1 to Technology 8, including a vibration sensor. With this, since the vibration detection mechanism includes a vibration sensor, vibrations propagating through at least one of the floor and the wall can be properly detected with a simple configuration.

[0166] (Technical 10) The vibration detection mechanism is an object detection system according to any one of Technical 1 to Technical 8, including a microphone. With this, since the vibration detection mechanism includes a microphone, vibrations propagating through at least one of the floor and the wall can be properly detected with a simple configuration.

[0167] As described above, the object detection system described herein has a simple configuration and can be used to accurately estimate at least one of the presence or absence of an object placed on the floor, the location of the object, and the type of the object.

[0168] 1 Object detection system 10 Lighting device (electrical equipment) 20 Switch (electrical equipment) 30A, 30B, 30C Outlet (electrical equipment) 4 Control device 41 Processor 411 Setting unit 412 Acquisition unit 413 Estimation unit 414 Notification unit 42 Memory 421 Control program 422 First model memory unit (model memory unit) 423 Second model memory unit (model memory unit) 424 Data memory unit 5, 5A, 5B, 5C, 5D Power line 50 Distribution board (electrical equipment) 6 Vibrator (multiple vibration application mechanisms) 24 First vibrator (vibration application mechanism) 34C Second vibrator (vibration application mechanism) 7 Sensor (multiple vibration detection mechanisms) 34A First sensor (vibration detection mechanism) 34B Second sensor (vibration detection mechanism) 8 Smartphone 9 Indoor space 91 Floor 93-96 Wall DK Desk DJ1 Input Dataset DQ1 Output Dataset DJ2 Input Dataset DQ2 Output Dataset FQ1 First Frequency FQ2 Second Frequency M Trained Model M1 First Trained Model (Trained Model) M2 Second Trained Model (Trained Model) P Human Body VA1, VA11, VA12 Applied Vibration Signals VA2, VA21, VA22 Applied Vibration Signals VB1, VB11, VB12 Detected Vibration Signals VB2, VB21, VB22 Detected Vibration Signals XD, YD, XP, YP Position

Claims

1. An object detection system comprising: a plurality of vibration application mechanisms for applying vibration to at least one of a floor and a wall; a plurality of vibration detection mechanisms for detecting vibrations propagating through at least one of the floor and the wall; and a control device for controlling the plurality of vibration application mechanisms, wherein the control device comprises: an estimation unit for estimating at least one of the presence or absence of an object placed on the floor, the position of the object, and the type of the object, based on the detection results of the plurality of vibration detection mechanisms; and a notification unit for notifying the estimation results of the estimation unit.

2. The object detection system according to claim 1, wherein the control device comprises a model storage unit that generates a trained model learned by machine learning using the detection results of the plurality of vibration detection mechanisms as an input dataset and at least one of the presence or absence of the object, the position of the object, and the type of the object as an output dataset, and stores the generated trained model, and the estimation unit estimates at least one of the presence or absence of the object, the position of the object, and the type of the object using the trained model.

3. The object detection system according to claim 1, wherein the control device includes a setting unit for each of the plurality of vibration application mechanisms, which sets the frequency of the vibration to be applied according to the type of object.

4. The object detection system according to any one of claims 1 to 3, wherein each of the plurality of vibration application mechanisms, the plurality of vibration detection mechanisms, and the control device is connected to a commercial power supply and is arranged on electrical equipment that is fixedly installed in a living space.

5. The object detection system according to claim 1, wherein the vibration application mechanism includes a vibration motor.

6. The object detection system according to claim 1, wherein the vibration application mechanism includes a piezoelectric element.

7. The object detection system according to claim 1, wherein the vibration application mechanism includes an electrostrictive vibrator.

8. The object detection system according to claim 1, wherein the vibration application mechanism includes a magnetostrictive oscillator.

9. The object detection system according to claim 1, wherein the vibration detection mechanism includes a vibration sensor.

10. The object detection system according to claim 1, wherein the vibration detection mechanism includes a microphone.