Seating detection device, seating detection method, and program

Abstract

A seating detection device (10) comprises: one or more receivers (12) that are arranged in a space having a plurality of seats and receive sound generated inside or outside the space; an acoustic characteristic analysis unit (13) that calculates the time characteristics or frequency characteristics of the sound in the space from the signals received by the one or more receivers (12); and a detection unit (14) that detects the presence or absence of personnel or seating in the space on the basis of the time characteristics or frequency characteristics calculated by the acoustic characteristic analysis unit (13), and outputs the detection results.

Description

Technical Field

[0005] ,

[0001] The present disclosure relates to a technology for detecting the presence or absence of a person or the presence or absence of seating.

Background Art

[0002] Conventionally, technologies for detecting the presence or absence of a person or the presence or absence of seating have been known (see, for example, Patent Documents 1 to 4).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the technologies described in Patent Documents 1 and 2, there is a problem that sensors need to be embedded for each seat to be detected. Further, in the technology described in Patent Document 3, there is a problem that a set of sensors can only detect the presence or absence of seating of one seat. Further, in the technology described in Patent Document 4, since image processing using a camera needs to be performed, there is a problem that the memory load is relatively high.

[0005] Therefore, the present disclosure has been made in view of the above problems, and it is possible to detect the presence or absence of a person or the presence or absence of seating of two or more seats with a single set of sensors without the need to embed a sensor for each seat to be detected, and an object of the present disclosure is to provide a seating detection device that does not require image processing using a camera.

Means for Solving the Problems

[0006] A seating detection device according to one aspect of the present disclosure comprises: one or more receivers arranged in a space having multiple seats, which receive sounds generated in or outside the space; an acoustic characteristics analysis unit which calculates the temporal characteristics or frequency characteristics of the sound in the space from the signals received by the one or more receivers; and a detection unit which detects the presence or absence of a person or seated person in the space based on the temporal characteristics or frequency characteristics calculated by the acoustic characteristics analysis unit and outputs the detection result.

[0007] A seating detection method according to one aspect of the present disclosure is a seating detection method performed by a seating detection device comprising one transmitter, one or more receivers, an acoustic characteristics analysis unit, and a detection unit, wherein the one transmitter sequentially transmits a predetermined sound into space, the one or more receivers sequentially receive the reflected waves of the predetermined sound transmitted by the one transmitter, and the acoustic characteristics analysis unit analyzes a transmission signal indicating the waveform of the predetermined sound transmitted by the one transmitter and one or more received signals indicating the waveforms of the reflected waves of the predetermined sound received by the one or more receivers. The detection unit sequentially calculates the temporal or frequency characteristics of the sound in the vehicle interior space caused by the predetermined sound from at least one of the following: the acoustic characteristics analysis unit calculates the temporal or frequency characteristics of the sound in the vehicle interior space caused by the predetermined sound transmitted by the transmitter at a first timing, and the temporal or frequency characteristics of the sound in the vehicle interior space caused by the predetermined sound transmitted by the transmitter at a second timing, and detects the presence or absence of a person or seated person in the space, and outputs the detection result.

[0008] A program according to one aspect of this disclosure is a program for causing a seating detection device, comprising one transmitter, one or more receivers, an acoustic characteristics analysis unit, and a detection unit, to perform seating detection processing, wherein the seating detection processing involves the one transmitter sequentially transmitting a predetermined sound into space, the one or more receivers sequentially receiving reflected waves of the predetermined sound transmitted by the one transmitter, and the acoustic characteristics analysis unit analyzing one or more waveforms of the reflected waves of the predetermined sound received by the one or more receivers, including a transmission signal showing the waveform of the predetermined sound transmitted by the one transmitter and one or more waveforms of the reflected waves of the predetermined sound received by the one or more receivers. The process includes sequentially calculating the temporal or frequency characteristics of the sound in the vehicle interior space due to the predetermined sound from at least one of the received signals, and the detection unit detecting the presence or absence of people or seated persons in the space based on the difference between the temporal or frequency characteristics of the sound in the vehicle interior space due to the predetermined sound transmitted by the transmitter at a first timing and the temporal or frequency characteristics of the sound in the vehicle interior space due to the predetermined sound transmitted by the transmitter at a second timing, as calculated by the acoustic characteristics analysis unit, and outputting the detection result. [Effects of the Invention]

[0009] According to the seating detection device, seating detection method, and program described herein, there is no need to embed sensors in each seat to be detected, and a single set of sensors can detect the presence or absence of people or whether people are seated in two or more seats. Furthermore, a seating detection device is provided that does not require image processing using a camera. [Brief explanation of the drawing]

[0010] [Figure 1] Figure 1 is a block diagram showing the configuration of the seating detection device according to Embodiment 1. [Figure 2] Figure 2 is a schematic plan view showing how the seating detection device according to Embodiment 1 is installed inside a vehicle. [Figure 3] Figure 3 is a waveform diagram showing an example of a transmission signal according to Embodiment 1. [Figure 4]Figure 4 is a waveform diagram showing an example of an impulse response generated by the acoustic characteristics analysis unit according to Embodiment 1. [Figure 5] Figure 5 is a block diagram showing the configuration of the detection unit according to Embodiment 1. [Figure 6] Figure 6 is a flowchart of the first seating detection process according to Embodiment 1. [Figure 7] Figure 7 is a block diagram showing the configuration of the seat detection device according to Embodiment 2. [Figure 8] Figure 8 is a block diagram showing the configuration of the seating detection device according to Embodiment 3. [Figure 9] Figure 9 is a block diagram showing the configuration of the seating detection device according to Embodiment 4. [Figure 10] Figure 10 is a schematic plan view showing a vehicle equipped with a seating detection device according to Embodiment 4. [Figure 11] Figure 11 is a schematic diagram showing an example of the correspondence between the learning model according to Embodiment 4, the training data used for its training, the conditions that the training data satisfies, and the correct labels assigned to the training data. [Figure 12] Figure 12 is a flowchart of the second seating detection process according to Embodiment 4. [Modes for carrying out the invention]

[0011] (The circumstances that led to obtaining one aspect of this disclosure) Through the study of a seating detection device that detects the presence or absence of seating in a space having a plurality of seats, when sequentially calculating an impulse response in the space by ultrasonic waves using a transmitter and a receiver of ultrasonic waves installed in the space, the inventors obtained the finding that a difference occurs in the calculated impulse response when there is human movement. More specifically, when there is human movement between the first timing and the second timing, even if the movement is a relatively small movement such as the movement of a person accompanying breathing, the inventors obtained the finding that a difference occurs between the first impulse response by a predetermined ultrasonic wave transmitted at the first timing by the transmitter and the second impulse response by a predetermined ultrasonic wave transmitted at the second timing by the transmitter. Then, based on the above findings, the inventors further repeated the study and came up with the following seating detection device and the like that do not require embedding sensors for each seat to be detected and can detect the presence or absence of personnel or the presence or absence of seating for two or more seats with a single set of sensors and do not require image processing using a camera.

[0012] A seating detection device according to an aspect of the present disclosure is disposed in a space having a plurality of seats, and includes one or more receivers that receive sound generated inside or outside the space, an acoustic characteristic analysis unit that calculates the temporal characteristic or frequency characteristic of the sound in the space from the signal received by the one or more receivers, and a detection unit that detects the presence or absence of personnel or the presence or absence of seating in the space based on the temporal characteristic or the frequency characteristic calculated by the acoustic characteristic analysis unit and outputs a detection result.

[0013] In the above seating detection device, the presence or absence of personnel or the presence or absence of seating in the space is reflected in the temporal characteristic or frequency characteristic of the sound in the space having a plurality of seats.

[0014] Therefore, according to the above seating detection device, it is not necessary to embed sensors for each seat to be detected, and it is possible to detect the presence or absence of personnel or the presence or absence of seating for two or more seats with a single set of sensors, and a seating detection device that does not require image processing using a camera is provided.

[0015] Furthermore, it further includes one transmitter for transmitting a predetermined sound into the space, the one or more receivers receive a reflected wave of the predetermined sound, and the acoustic characteristic analysis unit calculates the time characteristic or the frequency characteristic of the predetermined sound from the signal of the predetermined sound and the signal received by the one or more receivers. The detection unit detects the presence or absence of the person or the presence or absence of sitting based on the difference between the time characteristic or the frequency characteristic of the predetermined sound transmitted at the first timing by the one transmitter and the time characteristic or the frequency characteristic of the predetermined sound transmitted at the second timing by the one transmitter calculated by the acoustic characteristic analysis unit.

[0016] In the above seating detection device, when there is human movement between the first timing and the second timing, even if the movement is relatively minute movement such as human movement accompanying breathing, etc., there will be a difference between the time characteristic or the frequency characteristic corresponding to the first timing calculated by the acoustic characteristic analysis unit and the time characteristic or the frequency characteristic corresponding to the second timing.

[0017] Therefore, according to the above seating detection device, the presence or absence of a person or the presence or absence of sitting can be detected more accurately.

[0018] Also, the sound transmitted by the one transmitter may be ultrasonic wave.

[0019] Thereby, it is possible to suppress the discomfort that may be caused to the people in the space by the sound transmitted by the one transmitter.

[0020] Also, the one transmitter may be located at a place where it is possible to transmit sound to all seats in the space.

[0021] Thereby, it is possible to detect the presence or absence of a person or the presence or absence of sitting for all seats in the space.

[0022] Furthermore, the space is the interior space of the vehicle, and at least one of the one transmitter and the one or more receivers is located at a location offset from the centerline of the vehicle extending in the direction of travel, and the one transmitter is not located at a location symmetric to at least one of the one or more receivers with respect to the centerline as the axis of symmetry.

[0023] This allows at least one receiver to have different timings for receiving reflected waves from a person seated in one of two seats located symmetrically with respect to the center line, and for receiving reflected waves from a person seated in the other seat. Therefore, it is possible to distinguish between one of the two seats and detect whether or not it is occupied. In particular, when the space is a vehicle, as a result, in a vehicle where the seat arrangement is symmetrically installed, it is possible to detect the presence or absence of a person or whether it is occupied in all seats.

[0024] Furthermore, the space may be a vehicle, and the one transmitter and the one or more receivers may be located within the overhead console of the vehicle.

[0025] As a result, the seating detection device is positioned to have a view of all seats, allowing it to more effectively detect the presence or absence of people, or whether they are seated.

[0026] Furthermore, the one or more receivers are multiple receivers, and the acoustic characteristic analysis unit calculates the time characteristic or the frequency characteristic by performing directional control using at least two of the multiple received signals that represent the waveforms of the reflected waves received by the multiple receivers.

[0027] This makes it possible to distinguish the location of multiple seats and detect whether or not there is a person present or seated.

[0028] Furthermore, if the detection unit finds a difference of a predetermined value or more between the time characteristics or frequency characteristics of the predetermined sound transmitted by one transmitter at a first timing, calculated by the acoustic characteristics analysis unit, and the time characteristics or frequency characteristics of the predetermined sound transmitted by the transmitter at a second timing, the detection unit may calculate the seating position based on the position on the time axis in the time characteristics or frequency characteristics where the difference occurs.

[0029] This allows the seating position to be detected.

[0030] Furthermore, the one transmitter and at least one of the one or more receivers may be a speaker and a microphone used in a vehicle emergency call system, respectively.

[0031] This allows for a reduction in the number of speakers and microphones installed in the vehicle.

[0032] Furthermore, the detection unit may use a communication device used in the vehicle emergency call system to transmit the detected information regarding the presence or absence of a seat in the space to an emergency call center linked with the vehicle emergency call system.

[0033] This allows information regarding whether a vehicle is occupied or not to be transmitted to an emergency call center in the event of an emergency, such as a traffic accident, enabling more effective emergency medical services.

[0034] A seating detection method according to one aspect of the present disclosure is a seating detection method performed by a seating detection device comprising one transmitter, one or more receivers, an acoustic characteristics analysis unit, and a detection unit, wherein the one transmitter sequentially transmits a predetermined sound into space, the one or more receivers sequentially receive the reflected waves of the predetermined sound transmitted by the one transmitter, and the acoustic characteristics analysis unit analyzes a transmission signal indicating the waveform of the predetermined sound transmitted by the one transmitter and one or more received signals indicating the waveforms of the reflected waves of the predetermined sound received by the one or more receivers. The detection unit sequentially calculates the temporal or frequency characteristics of the sound in the vehicle interior space caused by the predetermined sound from at least one of the following: the acoustic characteristics analysis unit calculates the temporal or frequency characteristics of the sound in the vehicle interior space caused by the predetermined sound transmitted by the transmitter at a first timing, and the temporal or frequency characteristics of the sound in the vehicle interior space caused by the predetermined sound transmitted by the transmitter at a second timing, and detects the presence or absence of a person or seated person in the space, and outputs the detection result.

[0035] In a seating detection device that performs the above-described seating detection method, the temporal characteristics or frequency characteristics of sound in a space with multiple seats reflect the presence or absence of people in the space or whether or not they are seated.

[0036] Therefore, according to the above seating detection method, there is no need to embed a sensor in each seat to be detected, and it is possible to detect the presence or absence of people or whether they are seated in two or more seats with one set of sensors, and a seating detection device is provided that does not require image processing using a camera.

[0037] A program according to one aspect of this disclosure is a program for causing a seating detection device, comprising one transmitter, one or more receivers, an acoustic characteristics analysis unit, and a detection unit, to perform seating detection processing, wherein the seating detection processing involves the one transmitter sequentially transmitting a predetermined sound into space, the one or more receivers sequentially receiving reflected waves of the predetermined sound transmitted by the one transmitter, and the acoustic characteristics analysis unit analyzing one or more waveforms of the reflected waves of the predetermined sound received by the one or more receivers, including a transmission signal showing the waveform of the predetermined sound transmitted by the one transmitter and one or more waveforms of the reflected waves of the predetermined sound received by the one or more receivers. The process includes sequentially calculating the temporal or frequency characteristics of the sound in the vehicle interior space due to the predetermined sound from at least one of the received signals, and the detection unit detecting the presence or absence of people or seated persons in the space based on the difference between the temporal or frequency characteristics of the sound in the vehicle interior space due to the predetermined sound transmitted by the transmitter at a first timing and the temporal or frequency characteristics of the sound in the vehicle interior space due to the predetermined sound transmitted by the transmitter at a second timing, as calculated by the acoustic characteristics analysis unit, and outputting the detection result.

[0038] In a seat detection device that performs seat detection processing using the above program, the temporal characteristics or frequency characteristics of sound in a space with multiple seats reflect the presence or absence of people in the space or whether or not they are seated.

[0039] Therefore, according to the above program, there is no need to embed a sensor in each seat to be detected, and one set of sensors can detect the presence or absence of people or whether they are seated in two or more seats, and a seating detection device is provided that does not require image processing using a camera.

[0040] The following describes a specific example of a seating detection device according to one aspect of this disclosure, with reference to the drawings. The embodiments shown here are all specific examples of this disclosure. The numerical values, shapes, components, arrangement and connection configurations of components, as well as the steps (processes) and the order of steps shown in the following embodiments are examples and are not intended to limit this disclosure. Furthermore, each figure is a schematic diagram and is not necessarily a strict representation.

[0041] Furthermore, comprehensive or specific embodiments of this disclosure may be implemented as a system, method, integrated circuit, computer program, or recording medium such as a computer-readable CD-ROM, or as any combination of a system, method, integrated circuit, computer program, and recording medium.

[0042] (Embodiment 1) The seating detection device according to Embodiment 1 will be described below with reference to the drawings. This seating detection device is installed in a space with multiple seats and detects the presence or absence of people in the space or whether or not they are seated. The space with multiple seats may be, for example, a passenger car, a multi-passenger vehicle, a bus, a train, an aircraft, or a room. In this embodiment, as an example, the space with multiple seats will be described as a passenger car. However, the space with multiple seats is not limited to a passenger car.

[0043] <Structure> Figure 1 is a block diagram showing the configuration of the seating detection device 10 according to Embodiment 1.

[0044] As shown in Figure 1, the seat detection device 10 comprises a transmitter 11, a receiver 12, an acoustic characteristics analysis unit 13, a detection unit 14, an amplifier 15, an amplifier 16, an ADC 17 (ADC: Analog to Digital Converter), a DAC 18 (DAC: Digital to Analog Converter), and a transmission signal generation unit 19.

[0045] When an electrical signal is input to the transmitter 11, it converts the input electrical signal into sound and transmits the converted sound into the vehicle in which the seat detection device 10 is installed. In Embodiment 1, the sound converted by the transmitter 11, that is, the sound transmitted by the transmitter 11, will be described below as ultrasonic, but the sound transmitted by the transmitter is not necessarily limited to ultrasonic. The transmitter 11 can be implemented by, for example, a speaker. Alternatively, the transmitter 11 may be implemented by, for example, a piezoelectric element. The transmitter 11 transmits a predetermined ultrasonic sound when it receives a transmission signal, which will be described later.

[0046] The receiver 12 receives the sound transmitted by the transmitter 11 and its reflected waves, converts the received sound and its reflected waves into a received signal consisting of an analog electrical signal, and outputs the converted received signal. In Embodiment 1, the sound and its reflected waves received by the receiver 12 will be described below as ultrasonic waves, but the sound and its reflected waves received by the receiver 12 are not necessarily limited to ultrasonic waves. The receiver 12 may be implemented by, for example, a microphone. Alternatively, the receiver 12 may be implemented by, for example, a piezoelectric element. When the transmitter 11 and receiver 12 are implemented by piezoelectric elements, the transmitter 11 and receiver 12 may be implemented by a single piezoelectric element that operates in time-division multiplexing mode.

[0047] In this description, the seating detection device 10 is assumed to have one receiver 12, but the seating detection device 10 is not necessarily limited to a configuration with one receiver, and may have a configuration with multiple receivers.

[0048] Figure 2 is a schematic plan view showing how the seating detection device 10 is installed inside the vehicle 30. Figure 2 illustrates a part of the structure inside the vehicle 30 that cannot actually be directly seen, as if it were visible.

[0049] As shown in Figure 2, the seat detection device 10 is installed in an overhead console 32 located on the ceiling of the vehicle 30. More specifically, the seat detection device 10 is installed in the overhead console 32 at a location offset from the centerline of the vehicle 30 that extends in the direction of travel.

[0050] By installing the seating detection device 10 in the above position, the distance from which the ultrasonic waves transmitted from the transmitter 11 are reflected by the seat 31A and reach the receiver 12, the distance from which the ultrasonic waves transmitted from the transmitter 11 are reflected by the seat 31B and reach the receiver 12, the distance from which the ultrasonic waves transmitted from the transmitter 11 are reflected by the seat 31C and reach the receiver 12, and the distance from which the ultrasonic waves transmitted from the transmitter 11 are reflected by the seat 31D and reach the receiver 12 can be made to be different from each other.

[0051] Returning to Figure 1, let's continue the explanation of the seat detection device 10.

[0052] The transmission signal generation unit 19 generates a transmission signal that represents a predetermined ultrasonic waveform output by the transmitter 11, and outputs the generated transmission signal to the acoustic characteristic analysis unit 13 and the DAC 18. Here, the transmission signal generation unit 19 is described as generating a transmission signal consisting of digital signals. The transmission signal generation unit 19 is realized, for example, by a microprocessor (not shown) provided in the seat detection device 10 executing a program stored in the memory (not shown) provided in the seat detection device 10.

[0053] Figure 3 is a waveform diagram showing an example of a transmission signal output by the transmission signal generation unit 19.

[0054] In Figure 3, the horizontal axis represents time, the vertical axis in the waveform shown at the top represents the amplitude of the transmitted signal, and the vertical axis in the waveform shown at the bottom represents the frequency of the transmitted signal.

[0055] As shown in Figure 3, the transmitted signal is an 86ms signal consisting of a 43ms signaled period and a 43ms unsignaled period, both of which are sweep sine wave signals.

[0056] As shown in Figure 3, the transmission signal generation unit 19 continuously outputs a transmission signal consisting of a signaled period and a signalless period.

[0057] Returning to Figure 1, let's continue the explanation of the seat detection device 10.

[0058] The DAC18 converts the transmission signal, which is a digital signal generated by the transmission signal generation unit 19, into an analog signal, and outputs the transmission signal consisting of the converted analog signal to the amplifier 16.

[0059] The amplifier 16 amplifies the transmission signal, which has been converted into an analog signal by the DAC 18, and outputs the amplified transmission signal to the transmitter 11. As a result, the transmitter 11 transmits a predetermined ultrasonic wave. As described above, the transmission signal generation unit 19 outputs a continuous successive transmission signal. Therefore, the transmitter 11 continuously transmits the predetermined ultrasonic wave sequentially.

[0060] The amplifier 15 amplifies the received signal, which consists of the analog signal output from the receiver 12, and outputs the amplified received signal to the ADC 17.

[0061] The ADC17 converts the received signal, which consists of an amplified analog signal output from the amplifier 16, into a digital signal, and outputs the converted digital signal to the acoustic characteristics analysis unit 13.

[0062] As described above, the transmitter 11 continuously transmits predetermined ultrasonic waves. Therefore, the receiver 12 continuously receives the predetermined ultrasonic waves and their reflected waves. Consequently, the ADC 17 continuously outputs received signals corresponding to each of the predetermined ultrasonic waves transmitted from the transmitter 11.

[0063] The acoustic characteristics analysis unit 13 calculates the temporal or frequency characteristics of sound in the space inside the vehicle 30, that is, the space with multiple seats, from the signal received by the receiver 12. More specifically, the acoustic characteristics analysis unit 13 receives each of the transmission signals sequentially output from the transmission signal generation unit 19 and each of the reception signals sequentially output from the ADC 17 as input. For each of the sequentially input transmission signals, it sequentially calculates a predetermined ultrasonic impulse response inside the vehicle 30 from the transmission signal and the corresponding reception signal, and sequentially outputs the calculated impulse response to the detection unit 14.

[0064] Here, the impulse response of a predetermined ultrasonic wave within the vehicle 30 calculated by the acoustic characteristics analysis unit 13 is an example of the temporal or frequency characteristics of sound in the space within the vehicle 30. In the following explanation, the temporal or frequency characteristics of sound within the vehicle 30 calculated by the acoustic characteristics analysis unit 13 will be described as the impulse response of a predetermined ultrasonic wave within the vehicle 30. However, the temporal or frequency characteristics of sound within the vehicle 30 calculated by the acoustic characteristics analysis unit 13 are not necessarily limited to the impulse response of a predetermined ultrasonic wave within the vehicle 30. The acoustic characteristics analysis unit 13 is realized, for example, by a microprocessor (not shown) provided in the seat detection device 10 executing a program stored in the memory (not shown) provided in the seat detection device 10.

[0065] Figure 4 is a waveform diagram showing an example of the impulse response generated by the acoustic characteristics analysis unit 13.

[0066] In Figure 4, the horizontal axis represents time, and the vertical axis represents signal level.

[0067] As shown in Figure 4, the time-axis positions of the reflection component from seat 31A, seat 31B, seat 31C, and seat 31D in the impulse response are all different from each other. This is because, as mentioned above, the distance from which the ultrasonic waves transmitted from transmitter 11 are reflected by seat 31A and reach receiver 12 is different from the distance from which the ultrasonic waves transmitted from transmitter 11 are reflected by seat 31B and reach receiver 12 is different from the distance from which the ultrasonic waves transmitted from transmitter 11 are reflected by seat 31C and reach receiver 12 is different from the distance from which the ultrasonic waves transmitted from transmitter 11 are reflected by seat 31D and reach receiver 12.

[0068] Thus, the position on the time axis in the impulse response corresponds to the length of the path from the transmitter to the location where human movement occurs and then to the receiver.

[0069] The acoustic characteristics analysis unit 13 may, for example, perform a Fourier transform on the transmitted signal and the received signal, calculate the ratio of signal strength for each frequency for the transmitted signal and the received signal after the Fourier transform, and then calculate the impulse response by performing an inverse Fourier transform on the calculated ratio of signal strength for each frequency.

[0070] The detection unit 14 detects the presence or absence of people or seated persons in the vehicle 30 based on the temporal or frequency characteristics of the sound in the space inside the vehicle 30 calculated by the acoustic characteristics analysis unit 13, and outputs the detection result. More specifically, the detection unit 14 receives the impulse response sequentially calculated by the acoustic characteristics analysis unit 13 as input, and detects the presence or absence of people or seated persons in the vehicle 30 based on the difference between the impulse response of a predetermined ultrasonic wave transmitted by the transmitter 11 at a first timing (hereinafter also referred to as the "first impulse response") and the impulse response of a predetermined ultrasonic wave transmitted by the transmitter at a second timing (hereinafter also referred to as the "second impulse response"), and outputs the detection result. The detection unit 14 is realized, for example, by a microprocessor (not shown) provided in the seated detection device 10 executing a program stored in the memory (not shown) provided in the seated detection device 10.

[0071] Figure 5 is a block diagram showing the configuration of the detection unit 14. Schematic waveform diagrams of the signals output from the main components of the detection unit 14 are superimposed in Figure 5. In each waveform diagram, the horizontal axis represents time, and the vertical axis represents the signal level.

[0072] As shown in Figure 5, the detection unit 14 includes an impulse response holding unit 21, a difference calculation unit 22, an absolute value calculation unit 23, and a change position calculation unit 24.

[0073] The impulse response storage unit 21 stores each of the impulse responses that are sequentially calculated by the acoustic characteristic analysis unit 13.

[0074] When the difference calculation unit 22 receives a new impulse response from the acoustic characteristic analysis unit 13, it takes the newly input impulse response as the first impulse response and one of the impulse responses stored by the impulse response holding unit 21 (for example, the impulse response input immediately before the first impulse response) as the second impulse response, calculates the difference between these first and second impulse responses, and outputs a difference signal indicating the calculated difference to the absolute value calculation unit 23.

[0075] As shown in Figure 5, the time-axis positions of the reflection component from seat 31A, seat 31B, seat 31C, and seat 31D in the difference signal are all different from each other. Therefore, for example, if a person is seated in seat 31A, their movements (e.g., breathing) will cause a difference of a predetermined value or more in the time-axis position of the reflection component from seat 31A in the difference signal. The same applies if a person is seated in seat 31B, seat 31C, or seat 31D. Thus, when a person is seated in a seat in the vehicle 30, the position of the seat they are seated in is reflected in the time-axis position of the difference in the difference signal that exceeds a predetermined value.

[0076] When the absolute value calculation unit 23 receives a difference signal from the difference calculation unit 22, it calculates the absolute value of the input difference signal and outputs the calculated absolute value of the difference signal to the change position calculation unit 24.

[0077] When the change position calculation unit 24 receives an absolute value difference signal from the absolute value calculation unit 23, it detects seating within the vehicle 30 based on the difference absolute value signal and outputs detection information indicating the detection result. More specifically, the change position calculation unit 24 detects seating if a difference of a predetermined value or more occurs in the difference absolute value signal and outputs seating information indicating that seating is present. If no difference of a predetermined value or more occurs, it detects no seating and outputs seating information indicating that seating is absent.

[0078] The change position calculation unit 24 further calculates the seating position from the position on the time axis in the difference absolute value signal where the difference occurs if a difference of a predetermined value or more occurs in the difference absolute value signal, and includes information indicating the calculated seating position in the output seating information.

[0079] Furthermore, the change position calculation unit 24 is not limited to calculating seating information using only one difference absolute value signal; it may also utilize the average or maximum value of past difference absolute value signals, change frequency, etc. Alternatively, this information may be used as features and determined by machine learning.

[0080] If the decision is made using machine learning, it can be achieved as follows:

[0081] In the learning phase, a learning model is constructed using training data consisting of numerous differential absolute value signals, such as past differential absolute value signals, the average, maximum, and change frequencies of multiple differential absolute value signals, each labeled with whether or not a seat is occupied. The trained model is then output. In the utilization phase, at least one of the differential absolute value signals acquired by the seat detection device, the average, maximum, or change frequency of multiple differential absolute value signals, is input to the generated trained model. After internal processing of the trained model, it is determined whether or not a seat is occupied. The machine learning algorithm is not particularly limited as long as it achieves the output result described above. For example, it could be a commonly used supervised learning algorithm such as logistic regression or support vector machine, or it could be deep learning using a neural network that can discover features on its own.

[0082] <Operation> The seating detection device 10 configured as described above performs a first seating detection process to detect seating in the vehicle 30.

[0083] The first seating detection process performed by the seating detection device 10 will be explained below with reference to the drawings.

[0084] Figure 6 is a flowchart of the first seating detection process.

[0085] The first seating detection process is initiated, for example, when an impulse response previously calculated by the acoustic characteristics analysis unit 13 is stored in the impulse response holding unit 21, and a transmission signal is output from the transmission signal generation unit 19.

[0086] When the first seating detection process is initiated, the transmitter 11 transmits a predetermined ultrasonic wave into the vehicle 30 (step S10).

[0087] When a predetermined ultrasonic wave is transmitted from the transmitter 11, the receiver 12 receives the ultrasonic wave and its reflected waves, and outputs a received signal (step S20).

[0088] When a received signal is output from the receiver 12, the acoustic characteristic analysis unit 13 calculates the impulse response inside the vehicle 30 due to a predetermined ultrasonic wave from the transmitted signal output by the transmitted signal generation unit 19 and the received signal output by the receiver 12 (step S30), and outputs the calculated impulse response.

[0089] When an impulse response is output from the acoustic characteristics analysis unit 13, the difference calculation unit 22 uses this impulse response as the first impulse response and uses an impulse response previously calculated by the acoustic characteristics analysis unit 13 and stored in the impulse response holding unit 21 as the second impulse response. The difference between these first and second impulse responses is calculated and a difference signal indicating the calculated difference is output. The absolute value calculation unit 23 then calculates the absolute value of the difference signal output from the difference calculation unit 22 and outputs a difference absolute value signal indicating the calculated absolute value of the difference signal (step S40).

[0090] When the absolute value calculation unit 23 outputs a difference absolute value signal, the change position calculation unit 24 checks whether a difference of a predetermined value or more has occurred in that difference absolute value signal (step S50).

[0091] In the process of step S50, if a difference greater than a predetermined value occurs (step S50: Yes), the change position calculation unit 24 further calculates the seating position from the position on the time axis of the absolute difference signal in which the difference occurs (step S60). Then, the change position calculation unit 24 outputs seating information indicating that a seat has been found and the position of the seat (step S70).

[0092] In the process of step S50, if no difference greater than a predetermined value occurs (step S50: No), the change position calculation unit 24 outputs a seating signal indicating that there is no seating (step S80).

[0093] When the process in step S70 is completed, and when the process in step S80 is completed, the seating detection device 10 terminates its first seating detection process.

[0094] <Consideration> With the seating detection device 10 configured as described above, it is not necessary to embed sensors (here, transmitter 11 and receiver 12) in each seat to be detected (here, seats 31A to 31D). Furthermore, one set of sensors (here, transmitter 11 and receiver 12) can detect the presence or absence of a person or seating in two or more seats (here, seats 31A to 31D). In addition, since there is no need to perform image processing using a camera, it is possible to reduce the memory load and computational load compared to conventional devices that detect seating by performing image processing using a camera.

[0095] Furthermore, the seat detection device 10 calculates the impulse response inside the vehicle 30 using ultrasound. Generally, the noise inside a vehicle is less with ultrasound than with sound waves. For this reason, the seat detection device 10 can detect the presence or absence of a person or whether they are seated with higher accuracy than a seat detection device that calculates the impulse response inside the vehicle using sound waves.

[0096] As described above, the seating detection device 10 detects the presence or absence of a person or whether a person is seated based on the presence or absence of human movement. Therefore, it can detect the presence or absence of a person or whether a person is seated regardless of the person's shape or posture. Furthermore, it can suppress false detections of the presence or absence of a person or whether a person is seated due to stationary objects such as luggage placed on the seat.

[0097] As described above, the seating detection device 10 does not require sensors to be embedded in each seat to be detected. Therefore, it can detect the presence or absence of a person or whether someone is seated, regardless of seat deformation (for example, deformation of the seat shape from a chair shape to a mat shape) or the state in which parts of the seat are stored.

[0098] Generally, ultrasound can diffract and penetrate behind objects. Therefore, the seating detection device 10 can detect the presence or absence of a person or whether someone is seated even in areas that cannot be directly seen from its installation location.

[0099] The seat occupancy detection device 10 can, for example, output a calculated seat occupancy signal to the ECU (Electronic Control Unit) that controls the vehicle 30. In this way, the ECU that controls the vehicle 30 can, for example, prevent the airbag from unnecessarily deploying to an unoccupied seat. The ECU that controls the vehicle 30 can also, for example, turn off the seat heater to an unoccupied seat to reduce battery consumption. Furthermore, the ECU that controls the vehicle 30 can, for example, suppress unnecessary airflow from the air conditioner to an unoccupied seat.

[0100] (Embodiment 2) The following describes a seat detection device according to Embodiment 2, which is configured with some changes to the configuration of the seat detection device 10 according to Embodiment 1. This seat detection device is installed in a vehicle that is compatible with a vehicle emergency call system (hereinafter also referred to as "the vehicle according to Embodiment 2"), that is, a vehicle equipped with a speaker, microphone, and communication device used in the vehicle emergency call system. The following description focuses on the differences between the seat detection device according to Embodiment 2 and the seat detection device 10. The vehicle emergency call system is, for example, e-Call.

[0101] Figure 7 is a block diagram showing the configuration of the seating detection device 10A according to Embodiment 2.

[0102] As shown in Figure 7, the seating detection device 10A is configured in which the transmitter 11 is changed to a transmitter 51, the receiver 12 is changed to a receiver 52, and the detection unit 14 is changed to a detection unit 14A, compared to the seating detection device 10 according to Embodiment 1.

[0103] Transmitter 51 is a speaker used in the emergency call system and has the same function as transmitter 11. That is, transmitter 51 is used both as a transmitter for the seat detection device 10A and as a speaker used in the emergency call system.

[0104] Receiver 52 is a microphone used in the emergency call system and has the same functions as receiver 12. That is, receiver 52 is used both as a receiver for the seat detection device 10A and as a microphone used in the emergency call system.

[0105] In addition to the functions of the detection unit 14, the detection unit 14A has the following additional functions.

[0106] In other words, the additional function is a function that, when predetermined conditions are met, uses the communication device 50 used in the vehicle emergency call system to transmit to the emergency call center information regarding the presence or absence of people or whether they are seated inside the vehicle according to Embodiment 2, which has been detected. Here, the predetermined conditions may be, for example, conditions in which the emergency call system determines that the vehicle according to Embodiment 2 has been involved in an accident, or for example, conditions in which the emergency call system determines that a person riding in the vehicle according to Embodiment 2 has suddenly fallen ill.

[0107] <Consideration> With the seat detection device 10A configured as described above, the speaker and microphone are shared between the seat detection device 10A and the emergency call system. Therefore, the number of speakers and microphones installed in the vehicle according to Embodiment 2 can be reduced.

[0108] Furthermore, with the seat detection device 10A configured as described above, in the event of an accident involving the vehicle according to Embodiment 2, or in the event of an emergency such as a person in the vehicle according to Embodiment 2 suddenly becoming ill, information regarding the presence or absence of people or whether people are seated can be transmitted to the emergency call center. As a result, emergency medical personnel dispatched to the vehicle according to Embodiment 2 at the request of the emergency call center can know in advance which seats are occupied, how many people are in the vehicle, etc., thereby enabling faster and / or more reliable emergency medical services.

[0109] (Embodiment 3) The following describes a seat detection device according to Embodiment 3, which is configured with some changes to the configuration of the seat detection device 10 according to Embodiment 1. In the following, the seat detection device according to Embodiment 3 will be described, focusing on the differences from the seat detection device 10.

[0110] Figure 8 is a block diagram showing the configuration of the seating detection device 10B according to Embodiment 2.

[0111] As shown in Figure 8, the seat detection device 10B is configured as follows: the receiver 12 is changed to receivers 12A to 12N, the amplifier 15 is changed to amplifiers 15A to 15N, the ADC 17 is changed to ADC 17A to 17N, and a directional control unit 60 is added compared to the seat detection device 10 according to Embodiment 1.

[0112] Receivers 12A to 12N each have the same functions as receiver 12.

[0113] Amplifiers 15A to 15N each have the same function as amplifier 15.

[0114] ADC17A to ADC17N each have the same functions as ADC17.

[0115] In other words, the seat detection device 10B is configured by changing the number of receivers 12 from 1 to K, the number of amplifiers 15 from 1 to K, and the number of ADCs 17 from 1 to K. Here, although Figure 8 is illustrated as if K is an integer of 3 or more, K does not necessarily have to be an integer of 3 or more as shown in Figure 8, as long as it is an integer of 2 or more.

[0116] The directional control unit 60 receives K received signals output from each of the ADC17A to ADC17N, and performs directional control using at least two of the K received signals to calculate a received signal that emphasizes the reflected wave component from a specific direction. The calculated received signal, which emphasizes the reflected wave component from a specific direction, is then output to the transmission signal generation unit 19. The directional control unit 60 is realized, for example, by a microprocessor (not shown) in the seat detection device 10B executing a program stored in the memory (not shown) of the seat detection device 10B.

[0117] <Consideration> According to the seating detection device 10B with the above configuration, the seating position can be calculated with even greater accuracy than the seating detection device 10 according to Embodiment 1.

[0118] (Embodiment 4) The following describes a seating detection device according to Embodiment 4, which is configured by changing some of the components of the seating detection device 10 according to Embodiment 1.

[0119] In the following, we will describe the seating detection device according to Embodiment 4, focusing on the differences from the seating detection device 10.

[0120] Figure 9 is a block diagram showing the configuration of the seating detection device 10C according to Embodiment 4.

[0121] As shown in Figure 9, the seating detection device 10C is configured by removing the transmitter 11, amplifier 16, DAC 18, and transmission signal generation unit 19 from the seating detection device 10 according to Embodiment 1, adding a door sensor 20 and a condition-specific learning model storage unit 70, changing the acoustic characteristics analysis unit 13 to an acoustic characteristics analysis unit 13C, and changing the detection unit 14 to a detection unit 14C.

[0122] Figure 10 is a schematic plan view showing a vehicle 30C equipped with a seating detection device 10C. In Embodiment 4, the seating detection device 10C is installed on a vehicle 30C that has a left front door, a right front door, a left rear door, a right rear door, and a hatchback, as shown in Figure 10, and is described as detecting the presence or absence of a person or whether someone is seated by utilizing the sound produced when the left front door, right front door, left rear door, right rear door, or hatchback (hereinafter also simply referred to as "door") of the vehicle 30C closes. However, the seating detection device according to Embodiment 4 is not necessarily limited to being installed on a structure like the vehicle 30C, as long as it is installed on a structure that has the function of producing sound in a space with multiple seats, and the sound used to detect the presence or absence of a person or whether someone is seated is not necessarily limited to the sound produced when a door closes.

[0123] Furthermore, although the seat detection device 10C is described below assuming it is mounted on the vehicle 30C, it is not necessarily required that all components of the seat detection device 10C be mounted on the vehicle 30C, as long as at least the receiver 12 is mounted on the vehicle 30C.

[0124] The door sensor 20 detects the open / closed state of each of the following: the left front door, the right front door, the left rear door, the right rear door, and the hatchback.

[0125] The acoustic characteristics analysis unit 13C calculates the temporal or frequency characteristics of the sound in the space inside the vehicle 30C, that is, the space with multiple seats, from the signal received by the receiver 12. More specifically, the acoustic characteristics analysis unit 13C receives the detection result of the door open / closed state output from the door sensor 20 and the received signal output from the ADC 17 as input. From the detection result of the door open / closed state, it detects a door closing event and acquires the received signals before and after the door closing event. Then, the acoustic characteristics analysis unit 13C calculates the temporal or frequency characteristics of the sound in the space inside the vehicle 30C from the acquired received signals.

[0126] Here, a door closing event refers to an event in which at least one of the following doors—the left front door, the right front door, the left rear door, the right rear door, and the hatchback—changes from an open state to a closed state. A door closing event occurs, for example, when an occupant of vehicle 30C gets in or out of the vehicle.

[0127] The conditional learning model memory unit 70 stores multiple machine learning models (hereinafter also simply referred to as "learning models") that have been trained under different conditions.

[0128] Each of the multiple learning models is pre-trained using training data that includes the time characteristics or frequency characteristics of sound inside the vehicle 30C, output from the acoustic characteristics analysis unit 13C, each with a corresponding ground truth level.

[0129] Here, the conditions that are different from each other include, at a minimum, the condition that the left front door closing event occurs while all other doors are closed, the condition that the right front door closing event occurs while all other doors are closed, the condition that the left front door closing event occurs while all other doors are closed, and the condition that the hatchback door closing event occurs while all other doors are closed.

[0130] Furthermore, the multiple learning models stored in the condition-based learning model memory unit 70 include at least: learning model A71A, which is trained using training data that satisfies the condition that a door closing event for the left front door occurs while all other doors are closed; learning model B71B, which is trained using training data that satisfies the condition that a door closing event for the right front door occurs while all other doors are closed; learning model C71C, which is trained using training data that satisfies the condition that a door closing event for the left rear door occurs while all other doors are closed; and learning model N71N, which is trained using training data that satisfies the condition that a door closing event for the hatchback occurs while all other doors are closed.

[0131] Figure 11 is a schematic diagram showing an example of the correspondence between the training data used for training, the conditions that the training data satisfies, and the correct answer level assigned to the training data for each of the multiple learning models stored in the conditional learning model memory unit 70.

[0132] The detection unit 14C detects the presence or absence of people or seated persons in the vehicle 30C based on the temporal or frequency characteristics of the sound in the space inside the vehicle 30C calculated by the acoustic characteristics analysis unit 13C, and outputs the detection result. More specifically, the detection unit 14C is equipped with a classifier 14C, and by inputting the temporal or frequency characteristics of the sound in the space inside the vehicle 30C calculated by the acoustic characteristics analysis unit 13C and the door closing event detected by the acoustic characteristics analysis unit 13C into the classifier 14C, it obtains the presence or absence of people or seated persons in the vehicle 30 output from the classifier 14C, and outputs the obtained presence or absence of people or seated persons in the vehicle 30.

[0133] When the classifier 140 receives the temporal or frequency characteristics of the sound in the space inside the vehicle 30C, calculated by the acoustic characteristics analysis unit 13C, and the door closing vent detected by the acoustic characteristics analysis unit 13C, it obtains a learning model from the condition-specific learning model storage unit 70 that has been trained using training data that satisfies the conditions corresponding to the input door closing event. The classifier 14C then inputs the temporal or frequency characteristics of the sound in the space inside the vehicle 30C into the acquired learning model, obtains the presence or absence of a person or seated person in the vehicle 30 output from the learning model, and outputs the acquired presence or absence of a person or seated person in the vehicle 30.

[0134] <Operation> The seating detection device 10C with the above configuration performs a second seating detection process to detect seating in the vehicle 30C.

[0135] The second seating detection process performed by the seating detection device 10C will be explained below with reference to the drawings.

[0136] Figure 12 is a flowchart of the second seating detection process.

[0137] When the second seating detection process is initiated, the acoustic characteristics analysis unit 13C begins attempting to detect a door closing event (step S110).

[0138] In the process of step S110, when the acoustic characteristics analysis unit 13C detects a door closing event (after repeating step S110:No, then step S110:Yes), it checks whether all doors other than the door to which the detected door closing event is located are closed (step S120).

[0139] In the process of step S120, if all doors other than the door that is the target of the detected door closing event are closed (step S120: Yes), the acoustic characteristic analysis unit 13C acquires the received signals before and after the detected door closing event (step S130).

[0140] Once the received signals before and after the detected door closing event are acquired, the acoustic characteristics analysis unit 13C calculates the temporal or frequency characteristics of the sound in the space inside the vehicle 30C from the acquired received signals (step S140).

[0141] Once the temporal or frequency characteristics of the sound in the space inside the vehicle 30C are calculated, the detection unit 14C uses a learning model trained with training data that satisfies the conditions corresponding to the door closing event detected by the acoustic characteristics analysis unit 13C to obtain whether there are people in the vehicle 30 or whether they are seated (step S150).

[0142] In the process of step S120, if at least one door other than the door that is the target of the detected door closing event is not closed (step S120: No), or when the process of step S150 is completed, the seating detection device 10C terminates its second seating detection process.

[0143] <Consideration> According to the seating detection device 10C with the above configuration, it is possible to detect the presence or absence of a person or whether or not someone is seated by utilizing the sound produced when the door closes.

[0144] (supplement) Although a seating detection device according to one aspect of the present disclosure has been described above based on Embodiments 1 to 3, the present disclosure is not limited to these embodiments. Various modifications that a person skilled in the art could conceive of these embodiments, or forms constructed by combining components from different embodiments, may also be included within the scope of one or more aspects of the present disclosure, as long as they do not depart from the spirit of the present disclosure.

[0145] (1) In Embodiment 1, the seat detection device 10 was described as being installed in the overhead console 32 at a location offset from the centerline of the vehicle 30 extending in the direction of travel. However, the seat detection device 10 is not necessarily limited to being installed in the overhead console 32 at a location offset from the centerline of the vehicle 30 extending in the direction of travel, as long as it is in a position where ultrasonic waves transmitted from the transmitter 11 can be reflected by at least one of the seats 31A to 31D and reach the receiver 12.

[0146] The position of the seat detection device 10 is not necessarily limited, but it is desirable that it be located at a position where the distance from which the ultrasonic waves transmitted from the transmitter 11 are reflected by the seat 31A and reach the receiver 12 (hereinafter also referred to as the "first distance"), the distance from which the ultrasonic waves transmitted from the transmitter 11 are reflected by the seat 31B and reach the receiver 12 (hereinafter also referred to as the "second distance"), the distance from which the ultrasonic waves transmitted from the transmitter 11 are reflected by the seat 31C and reach the receiver 12 (hereinafter also referred to as the "third distance"), and the distance from which the ultrasonic waves transmitted from the transmitter 11 are reflected by the seat 31D and reach the receiver 12 (hereinafter also referred to as the "fourth distance") are all different from each other. Generally, seats in a vehicle are located in positions that are symmetrical with respect to the center line of the vehicle that extends in the direction of travel. Therefore, in order to make the first distance, the second distance, the third distance, and the fourth distance different from each other in the vehicle 30, this can be achieved, for example, by having at least one of the transmitter 11 and the receiver 12 located at a location offset from the center line extending in the direction of travel of the vehicle 30, and by having the transmitter 11 not located at a location symmetrical to the receiver 12 with respect to the center line as the axis of symmetry.

[0147] (2) In Embodiment 1, the difference calculation unit 22 was described as calculating the difference between the first and second impulse responses, with the newly input impulse response being the first impulse response and the impulse response input immediately before the first impulse response being the second impulse response. However, the first and second impulse responses are not necessarily limited to the newly input impulse response and the impulse response input immediately before it, as long as they are impulse responses of ultrasonic waves transmitted by the transmitter 11 at different timings. For example, the newly input impulse response may be used as the first impulse response, and the impulse response of the unmanned vehicle 30, which has been calculated in advance by the acoustic characteristic analysis unit 13, may be used as the second impulse response.

[0148] (3) In Embodiment 1, the transmission signal generated by the transmission signal generation unit 19 was described as a signal consisting of a sweep sine wave. However, the transmission signal is not limited to a signal consisting of a sweep sine wave. The transmission signal may be, for example, a signal consisting of white noise, or a signal consisting of band noise or a sine wave.

[0149] Furthermore, the signal-enabled period and the signal-dead period are not limited to 43ms; they may be of different lengths.

[0150] (4) In Embodiment 1, the detection unit 14 was described as being configured to include an absolute value calculation unit 23. However, as another example of the configuration of the detection unit 14, a configuration without an absolute value calculation unit 23 is also possible. In this configuration, the difference calculation unit 22 calculates the difference signal and sends it to the change position calculation unit 24. When the difference signal is input from the difference calculation unit 22, the change position calculation unit 24 detects seating in the vehicle 30 based on the difference signal and outputs detection information indicating the detection result. More specifically, the change position calculation unit 24 detects seating when the difference in the difference signal exceeds a predetermined value and outputs seating information indicating that seating is present. When the difference does not exceed a predetermined value, it detects no seating and outputs seating information indicating that no seating is present. The change position calculation unit 24 further calculates the seating position from the position on the time axis in the difference signal where the difference occurs if a difference of a predetermined value or more occurs in the difference signal, and includes information indicating the calculated seating position in the output seating information.

[0151] (5) In Embodiment 1, the receiver 12 may be, for example, a digital MEMS (Micro Electro Mechanical System) microphone. In this case, since the output of the receiver 12 is a digital signal, the amplifier 15 and ADC 17 are not required.

[0152] (6) In Embodiment 1, a passenger car was given as an example of a space having multiple seats, but the space having multiple seats is not limited to a passenger car and may be a mobile body other than a passenger car. The seating detection device 10 can also be applied to seating detection in mobile bodies other than passenger cars, such as multi-passenger vehicles, buses, trains, and aircraft. In multi-passenger vehicles, buses, trains, and aircraft, crew members currently patrol to check seating, but by applying the seating detection device 10, it is possible to reduce the number of patrols by crew members, thereby reducing the risk of infectious diseases such as coronavirus.

[0153] Furthermore, the space with multiple seats does not need to be limited to a mobile vehicle; it may be any other location, such as a movie theater or other indoor space with seating.

[0154] (7) In Embodiment 4, the seating detection device 10C was described as detecting the presence or absence of a person or whether a person is seated by utilizing the sound generated when the door of the vehicle 30C closes. However, the sound that the seating detection device 10C uses to detect the presence or absence of a person or whether a person is seated is not necessarily limited to the sound generated when the door of the vehicle 30C closes, as long as it is a sound generated inside or outside the vehicle 30C. For example, the sound that the seating detection device 10C uses to detect the presence or absence of a person or whether a person is seated may be the sound of the door mirror being folded in, or external noise from the vehicle 30C. Furthermore, the type of sound that the seating detection device 10C uses to detect the presence or absence of a person or whether a person is seated is not particularly limited, and may be, for example, an audible sound or an ultrasonic sound.

[0155] (8) One aspect of the present disclosure may be a seating detection method in which characteristic components included in the seating detection device are steps, not limited to the seating detection device according to Embodiments 1 to 4. Another aspect of the present disclosure may be a program that causes a device including a computer to execute each characteristic step included in the seating detection method. Another aspect of the present disclosure may be a computer-readable non-temporary recording medium on which such a program is recorded. [Industrial applicability]

[0156] This disclosure can be widely used in seating detection devices and the like that detect the presence or absence of people or whether they are seated in a space with multiple seats. [Explanation of symbols]

[0157] 10, 10A, 10B Seating detection device 11, 51 Transmitter 12, 12A, 12B, 12N, 52 Receivers 13, 13C Acoustic Characteristics Analysis Department 14, 14A, 14C detection unit 15, 15A, 15B, 15N, 16 Amplifiers 17, 17A, 17B, 17N ADC 18 DAC 19. Transmission signal generation unit 20 Door Sensors 21 Impulse response holding unit 22 Difference calculation part 23 Absolute Value Calculation Unit 24 Change Position Calculation Unit 30, 30C vehicles Seats 31A, 31B, 31C, 31D 32 Overhead Console 50 Notification device 60 Directivity control section 70 Conditional Learning Model Memory Unit 71A Learning Model A 71B Learning Model B 71C Learning Model C 71N Learning Model N 140 Classifiers

Claims

1. One or more receivers are arranged within the interior space of a vehicle having multiple seats and receive sounds generated within the interior space. A transmitter that transmits a predetermined sound into the internal space, An acoustic characteristics analysis unit that calculates the temporal characteristics or frequency characteristics of the sound in the internal space from the signals received by one or more receivers, The system includes a detection unit that detects the presence or absence of seating and the position of seating within the internal space based on the time characteristics or frequency characteristics calculated by the acoustic characteristics analysis unit, and outputs the detection result. At least one of the one transmitter and the one or more receivers is located at a position offset from the centerline of the vehicle extending in the direction of travel. The aforementioned transmitter is not located in a position symmetric to at least one of the one or more receivers with respect to the center line as the axis of symmetry. The one or more receivers receive the predetermined sound reflected waves. The acoustic characteristics analysis unit calculates the time characteristics or frequency characteristics of the predetermined sound from the predetermined sound signal and the signals received by one or more receivers. The detection unit detects whether or not someone is seated based on the difference between the time characteristics or frequency characteristics of the predetermined sound transmitted at a first timing by one transmitter and the time characteristics or frequency characteristics of the predetermined sound transmitted at a second timing by one transmitter, which are calculated by the acoustic characteristics analysis unit. If it detects that someone is seated, it calculates the position of the detected seat. The aforementioned one transmitter and at least one of the aforementioned one or more receivers are, respectively, a speaker and a microphone used in a vehicle emergency call system. Seating detection device.

2. One or more receivers are arranged within the interior space of a vehicle having multiple seats and receive sounds generated within the interior space. A transmitter that transmits a predetermined sound into the internal space, An acoustic characteristics analysis unit that calculates the temporal characteristics or frequency characteristics of the sound in the internal space from the signals received by one or more receivers, The system includes a detection unit that detects the presence or absence of seating and the position of seating within the internal space based on the time characteristics or frequency characteristics calculated by the acoustic characteristics analysis unit, and outputs the detection result. At least one of the one transmitter and the one or more receivers is located at a position offset from the centerline of the vehicle extending in the direction of travel. The aforementioned transmitter is not located in a position symmetric to at least one of the one or more receivers with respect to the center line as the axis of symmetry. The one or more receivers receive the predetermined sound reflected waves. The acoustic characteristics analysis unit calculates the time characteristics or frequency characteristics of the predetermined sound from the predetermined sound signal and the signals received by one or more receivers. The detection unit detects whether or not someone is seated based on the difference between the time characteristics or frequency characteristics of the predetermined sound transmitted at a first timing by one transmitter and the time characteristics or frequency characteristics of the predetermined sound transmitted at a second timing by one transmitter, which are calculated by the acoustic characteristics analysis unit. If it detects that someone is seated, it calculates the position of the detected seat. The detection unit uses a communication device used in the vehicle emergency call system to transmit the detected information regarding whether or not there is a seat in the interior space to an emergency call center linked to the vehicle emergency call system. Seating detection device.

3. The sound transmitted by the aforementioned transmitter is ultrasonic. A seating detection device according to claim 1 or claim 2.

4. The aforementioned transmitter is located in a position that allows it to transmit sound to all seats within the interior space. A seating detection device according to any one of claims 1 to 3.

5. The one transmitter and the one or more receivers are located inside the overhead console of the vehicle. A seating detection device according to claim 1 or claim 2.

6. The aforementioned one or more receivers are multiple receivers, The acoustic characteristics analysis unit calculates the time characteristics or frequency characteristics by performing directional control using at least two of the multiple received signals that represent the waveforms of the reflected waves received by the multiple receivers. A seating detection device according to any one of claims 1 to 5.

7. The detection unit further calculates the seating position based on the position on the time axis in the time characteristics or frequency characteristics where the difference occurs, if there is a difference of a predetermined value or more between the time characteristics or frequency characteristics of the predetermined sound transmitted at a first timing by one transmitter and the time characteristics or frequency characteristics of the predetermined sound transmitted at a second timing by the transmitter, as calculated by the acoustic characteristics analysis unit. A seating detection device according to any one of claims 1 to 5.

8. A seating detection method performed by a seating detection device comprising one transmitter, one or more receivers, an acoustic characteristics analysis unit, and a detection unit, At least one of the one transmitter and one or more receivers is located in a vehicle having multiple seats, offset from the centerline extending in the direction of travel. The aforementioned transmitter is not located in a position symmetric to at least one of the one or more receivers with respect to the center line as the axis of symmetry. The aforementioned transmitter sequentially transmits a predetermined sound into the interior space of the vehicle. The one or more receivers successively receive the reflected waves of the predetermined sound transmitted by the one transmitter. The acoustic characteristics analysis unit sequentially calculates the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound from at least one of the transmission signals showing the waveform of the predetermined sound transmitted by the one transmitter and one or more received signals showing the waveforms of the reflected waves of the predetermined sound received by one or more receivers. The detection unit detects the presence or absence of a seat in the internal space and the position of the seat based on the difference between the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound transmitted by the transmitter at a first timing, calculated by the acoustic characteristics analysis unit, and the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound transmitted by the transmitter at a second timing, and outputs the detection result. The aforementioned one transmitter and at least one of the aforementioned one or more receivers are, respectively, a speaker and a microphone used in a vehicle emergency call system. Seating detection method.

9. A seating detection method performed by a seating detection device comprising one transmitter, one or more receivers, an acoustic characteristics analysis unit, and a detection unit, At least one of the one transmitter and one or more receivers is located in a vehicle having multiple seats, offset from the centerline extending in the direction of travel. The aforementioned transmitter is not located in a position symmetric to at least one of the one or more receivers with respect to the center line as the axis of symmetry. The aforementioned transmitter sequentially transmits a predetermined sound into the interior space of the vehicle. The one or more receivers successively receive the reflected waves of the predetermined sound transmitted by the one transmitter. The acoustic characteristics analysis unit sequentially calculates the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound from at least one of the transmission signals showing the waveform of the predetermined sound transmitted by the one transmitter and one or more received signals showing the waveforms of the reflected waves of the predetermined sound received by one or more receivers. The detection unit detects the presence or absence of a seat in the internal space and the position of the seat based on the difference between the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound transmitted by the transmitter at a first timing, calculated by the acoustic characteristics analysis unit, and the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound transmitted by the transmitter at a second timing, and outputs the detection result. The detection unit uses a communication device used in the vehicle emergency call system to transmit the detected information regarding whether or not there is a seat in the interior space to an emergency call center linked to the vehicle emergency call system. Seating detection method.

10. Seating detection system comprising one transmitter, one or more receivers, an acoustic characteristics analysis unit, and a detection unit. A program for causing a device to perform seat detection processing, At least one of the one transmitter and one or more receivers is located in a vehicle having multiple seats, offset from the centerline extending in the direction of travel. The aforementioned transmitter is not located in a position symmetric to at least one of the one or more receivers with respect to the center line as the axis of symmetry. The aforementioned seating detection process is: The aforementioned transmitter sequentially transmits a predetermined sound into the interior space of the vehicle. The one or more receivers successively receive the reflected waves of the predetermined sound transmitted by the one transmitter. The acoustic characteristics analysis unit sequentially calculates the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound from the transmission signal showing the waveform of the predetermined sound transmitted by the one transmitter and at least one of the one or more received signals showing the waveforms of the reflected waves of the predetermined sound received by the one or more receivers. The detection unit includes a process that detects the presence or absence of a seat in the internal space and the position of the seat based on the difference between the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound transmitted by the transmitter at a first timing, calculated by the acoustic characteristics analysis unit, and the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound transmitted by the transmitter at a second timing, and outputs the detection result. The aforementioned one transmitter and at least one of the aforementioned one or more receivers are, respectively, a speaker and a microphone used in a vehicle emergency call system. program.

11. Seating detection system comprising one transmitter, one or more receivers, an acoustic characteristics analysis unit, and a detection unit. A program for causing a device to perform seat detection processing, At least one of the one transmitter and one or more receivers is located in a vehicle having multiple seats, offset from the centerline extending in the direction of travel. The aforementioned transmitter is not located in a position symmetric to at least one of the one or more receivers with respect to the center line as the axis of symmetry. The aforementioned seating detection process is: The aforementioned transmitter sequentially transmits a predetermined sound into the interior space of the vehicle. The one or more receivers successively receive the reflected waves of the predetermined sound transmitted by the one transmitter. The acoustic characteristics analysis unit sequentially calculates the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound from the transmission signal showing the waveform of the predetermined sound transmitted by the one transmitter and at least one of the one or more received signals showing the waveforms of the reflected waves of the predetermined sound received by the one or more receivers. The detection unit includes a process that detects the presence or absence of a seat in the internal space and the position of the seat based on the difference between the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound transmitted by the transmitter at a first timing, calculated by the acoustic characteristics analysis unit, and the time characteristics or frequency characteristics of the sound in the internal space caused by the predetermined sound transmitted by the transmitter at a second timing, and outputs the detection result. The detection unit uses a communication device used in the vehicle emergency call system to transmit the detected information regarding whether or not there is a seat in the interior space to an emergency call center linked to the vehicle emergency call system. program.

Images