JP2004227053A - Life monitor device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a life monitor device to be installed in a living space in which a person to be watched resides. <P>SOLUTION: The life monitor device is provided with a vibration detecting means 11 that is installed in the living space in which the person to be monitored resides and detects a vibration generated in the living space, an information extracting means 12 which extracts information on a vibration generated by the action of the person in the living space from the output by the vibration detecting means and a transmitting means 13 which transmits the information extracted by the information extracting means. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a life monitoring device that tracks the daily activities of an elderly person or a person requiring care using a sensor and detects an abnormality of the elderly or a person requiring care.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, a system has been proposed in which a sensor for detecting occurrence of an abnormality in life such as a person falling down in a room and being unable to move is installed in a living space.
[0003]
For example, there is a system such as a living monitor system disclosed in Patent Literature 1, which reports the usage status of an electric appliance such as an electric pot used everyday to the outside. In this case, it is determined that there is an abnormality in the subject when the electric appliances that should be used daily are not used for a long time.
[0004]
Also, a system has been proposed in which a sensor is installed in the home of the target person to determine the behavior of the target person in the home. For example, in the living activity monitoring system disclosed in Patent Document 2, a mat equipped with a pressure sensor is arranged at a plurality of places such as an entrance of a toilet or a washroom, and when a subject steps on the mat, a signal is output, and the subject is output. Detect where you are and how it is moving.
[0005]
Furthermore, there is a system in which a dynamic detection sensor using light such as a photoelectric switch or an infrared sensor is installed on a wall or the like, such as a lifestyle abnormality detection system disclosed in Patent Document 3. That is, each time the subject passes through the sensing target area, the movement detection sensor outputs a sensor signal. Then, it is determined from the sensor signal that the subject is active in the home, and when the output of the sensor cannot be obtained throughout the day or when there is a difference from a normal behavior pattern, it is determined that the subject is abnormal. Automatically report to emergency call centers.
[0006]
[Patent Document 1]
JP-A-11-220779
[Patent Document 2]
JP-A-2000-57462
[Patent Document 3]
JP-A-11-9559
[0007]
[Problems to be solved by the invention]
However, in the case of notifying the use status of the electric appliance that is used daily as in Patent Literature 1, it is difficult to determine the abnormality because the time zone used by the subject does not always match every day. In addition, in the case of an electric kettle, there is a problem in that, if the subject to be used every morning has an abnormality in the daytime, it cannot be determined that the abnormality is until the next morning.
[0008]
On the other hand, in the case of the mat type sensor disclosed in Patent Literature 2, it is necessary to install the mat type sensor at a position where the subject must step on the sensor, and to reliably detect the sensor, the sensor must be arranged in a wide area. However, the number of sensors and the size of the mat are enlarged. Furthermore, there is a problem that it cannot be determined that a load or other object such as luggage is placed on the mat.
[0009]
Also, in the dynamic detection sensor using light disclosed in Patent Document 3, it is necessary to adjust the installation height and sensitivity in order to avoid erroneous detection by animals such as pets and the influence of sunlight, which makes installation work very difficult. There is a problem that is.
[0010]
Furthermore, in the devices according to Patent Documents 1 to 3, since the personal information of the target person cannot be extracted, the target person cannot be specified. Therefore, this device can be introduced only to the target person (single person) house living alone. For this reason, for example, when targeting the elderly living with the family, there is a problem that it is not possible to set up a system for notifying the family when an abnormality occurs while the family is away.
[0011]
The present invention has been made in view of such circumstances, and has as its object to specify a target person based on information extracted from the daily life of the target person, determine whether or not the target person is abnormal, and to simplify the method. It is an object of the present invention to provide a life monitoring device that can be installed in a home.
[0012]
[Means for Solving the Problems]
The life monitoring device of the present invention is installed in a living space where a target person (for example, a resident 50 to be described later) lives, and detects vibration generated in the living space; It is characterized by comprising information extracting means for extracting information relating to vibration generated by human motion in the living space from the output, and transmitting means for transmitting the information extracted by the information extracting means.
[0013]
A second aspect of the present invention is characterized in that the walking of the subject is specified from information on vibration generated by human motion.
[0014]
A third aspect of the present invention is an abnormality determining means for determining an abnormal state of the subject by comparing the extracted information with a predetermined determination criterion, and determining a result of the abnormality determining means in a predetermined predetermined condition. Communication means for notifying the report destination.
[0015]
In a fourth aspect of the present invention, the (vibration detecting means and the information extracting means) are respectively installed in the living space of the subject, and the information extracted by the information extracting unit includes each of the plurality of information extracting means. And an ID assigning means for assigning an ID (identification signal or identification information) corresponding to the information, and the transmitting means transmits the ID together with the information extracted by the information extracting means.
[0016]
A fifth aspect of the present invention is characterized in that the fifth aspect of the present invention is provided with an analyzing means for calculating an activity amount of the subject from information extracted from a plurality of information extracting means provided in the living space of the subject.
[0017]
[Action and effect]
According to the present invention, since information relating to vibration generated by human motion is extracted from the output of the vibration detecting means, even if vibration based on motion of an animal other than a human (such as a cat) is detected, The information based on the detection output is not extracted, and only the information regarding the vibration generated by the human motion can be transmitted.
[0018]
In the second aspect of the present invention, since the walking of the subject is specified from the information on the vibration generated by the motion of the human, only the behavior of the subject in the living space can be extracted, and the subject lives together with the family other than the subject. It can be applied to homes where people are living, facilities for the elderly with a plurality of subjects, etc., and a great effect can be obtained in practical use.
[0019]
In the third aspect of the present invention, by providing the abnormality determination means and the communication means, even when the target person becomes abnormal, it is not necessary to report by himself, and the abnormal state is designated in advance such as a family or a call center. Since a notification is sent to a predetermined notification destination, a prompt response can be taken.
[0020]
According to the fourth aspect of the present invention, since the ID transmitted together with the information extracted by the information extracting unit corresponds to the unit that has detected the vibration based on the information, the location where the vibration is generated Thus, the location of the subject is known. Therefore, even when the target person cannot move due to a fall or the like, prompt measures can be taken when a caregiver or a rescue worker rushes. In addition, the history of the resident's movement in the house can be grasped.
[0021]
In the fifth aspect of the present invention, since the activity amount of the subject is calculated from the vibration generated by the human motion at a plurality of locations, the daily activity amount can be grasped, and a physical condition abnormality or the like is predicted from the change in the activity amount. It is possible to achieve a great effect practically.
[0022]
BEST MODE FOR CARRYING OUT THE INVENTION
[First embodiment]
FIG. 1 shows a configuration of a life watching apparatus according to the first embodiment. This life watching device is configured to include a behavior mode extraction unit 1. The behavior mode extracting unit 1 detects vibration generated in a living space and outputs a vibration signal, and extracts information (for example, walking information to be described later) on vibration generated by human motion. The apparatus includes an information extracting unit 12 for outputting the information and a transmitting unit 13 for transmitting the information to the outside (for example, an abnormality determining unit 2 described later).
[0023]
FIG. 2 shows a configuration of the behavior mode extraction unit 1 and the abnormality determination unit 2. The abnormality determination unit 2 includes a reception unit 21 connected to the transmission unit 13 of the behavior mode extraction unit 1 by wireless or wire, an abnormality determination unit 22 that determines an abnormal state of the subject based on the above-described information, The communication means 23 includes a communication means 23 for notifying (transmitting) the state to the outside when the state is determined. The communication means 23 is connected to a network (for example, the Internet or a telephone line) 3.
[0024]
FIG. 3 shows a mode in which n behavior mode extraction units 1 are provided (n is an integer of 2 or more).
[0025]
The n behavior mode extraction units 1a to 1k each include vibration detection units 11a to 11k, information extraction units 12a to 12k connected downstream of the vibration detection units 11a to 11k, and transmission units 13a to 13k. Have been. The configuration of each vibration detection unit (for example, the vibration detection unit 11a) will be described in detail with reference to FIG. 4, and the configuration of each information extraction unit (for example, the information extraction unit 12a) will be described in detail with reference to FIG.
[0026]
FIG. 4 is a diagram showing the internal configuration of the vibration detecting means 11a.
[0027]
The vibration detecting unit 11a includes a vibration sensor 111a and a signal processing unit 112a. The signal processing unit 112a includes an amplifier 1112a, a filter 2112a, and an A / D converter 3112a. The vibration sensor 111a detects vibration of a floor, a wall, or the like, and outputs a vibration signal A according to the vibration. Specifically, an acceleration sensor, a speedometer, a microphone, a pressure gauge, a strain gauge, or the like is used as the vibration sensor 111a. For example, in the case of an acceleration sensor, if the vibration sensor 111a detects acceleration in one axis direction, the vibration sensor 111a installs a sensor surface on the floor and detects vertical vibration generated on the floor surface. The acceleration in the vertical direction may be measured using a three-axis acceleration sensor to detect the vibration with higher accuracy. The amplifier 1112a amplifies the vibration signal A, and the filter (for example, a low-pass filter) 2112a cuts the noise of the amplified vibration signal A. Since the manner of transmission of the vibration varies depending on each place in the living space, the state of the vibration detected by the vibration detecting means 11 is grasped at the time of installation, and an amplifier 1112a is provided so that a vibration signal can be optimally obtained for each installation place. It is also conceivable to automatically adjust the amplification factor of. The filter 2112a selects a noise cut frequency in accordance with vibration generated by a human. For example, when extracting mainly information about human walking, a low-pass filter set to a cut frequency of several hundred Hz may be used. The A / D converter 3112a converts the noise-cut vibration signal A from an analog signal to a digital signal. Note that the A / D converter 3112a may analyze the frequency of the vibration signal A, extract which frequency component signal is strong, and digitize the signal. Then, the result B converted into a digital signal is sent to the information extracting means 12a.
[0028]
FIG. 5 shows a mode in which the vibration detecting unit 11a of the behavior mode extracting unit 1a detects a vibration generated by a resident walking.
[0029]
The resident 50 walks around the living space for daily life. At that time, the right foot 50a and the left foot 50b are alternately separated from the floor 30 and proceed to the destination. Specifically, the left and right feet 50a and 50b alternately perform an operation in which the toes of the left foot 50b leave the floor 30 and the heels of the left foot 50b land on the floor 30. At the time of this landing, the floor 30 vibrates, and the vibration signal 51 transmitted to the floor is detected by the vibration detecting means 11a in the behavior mode extracting unit 1a. Here, unlike a conventional mat-type sensor or a sensor using light, it is only necessary to bring the vibration detecting means 11a in the behavior mode extraction unit 1a into contact with the floor on which vibration is transmitted. Since the action mode extraction unit 1a may be installed anywhere, it is possible to detect the action of the target person without any trouble of installing the sensor and without considering the detection range like a mat.
[0030]
FIG. 6 shows a waveform of a vibration signal A obtained by detecting the vibration signal 51 transmitted to the floor by the vibration sensor 111a inside the vibration detection means 11a and a waveform of a signal output B after signal processing by the signal processing unit 112a. FIG.
[0031]
FIG. 6A shows the waveform of the vibration signal A with time (t) on the horizontal axis and voltage (V) on the vertical axis. This waveform includes a first waveform 51a (amplitude indicates a maximum value and a minimum value) generated when the heel or toe of the foot of the resident 50 lands on the floor 30, and a second waveform as a reverberation of the first waveform 51a. 51b. In FIG. 6A, since the resident 50 is walking in a direction away from the vibration detection means 11, the amplitude of the vibration signal is gradually attenuated.
[0032]
FIG. 6B shows the output timing of the signal of the A / D converter 3112a at the point B, which is output after a predetermined time ta from the first waveform 51a of the vibration. The output signal of the A / D converter 3112a includes a content indicating that vibration has occurred, an analysis result of a frequency component of the signal, and the like. Note that the time ta is delayed from the first waveform 51a of the vibration signal A in order to analyze the frequency component, and is about several hundred ms. However, if the frequency analysis is not performed, the first waveform 51a May be output after detection.
[0033]
FIG. 7 is a diagram illustrating an analysis result of a frequency component of the vibration signal A extracted by the signal processing unit 112a. FIG. 7A shows an example when the object is dropped on the floor, and FIGS. 7B and 7C show an example when a human walks. The horizontal axis represents the frequency, and the vertical axis represents the spectrum intensity of the signal. A strong spectrum is detected only at the frequency f1 in the vibration spectrum when the object (1) falls. On the other hand, when a human is walking, spectra are detected at a plurality of frequencies as shown in (2) and (3). Further, as shown in (2) and (3), the frequency at which the spectrum is strong differs depending on the difference in footwear, such as how people walk and walk, and slippers. For example, characteristics of an individual can be extracted, such as a person who walks by kicking strongly with his toes after landing with his heels, and a person who lands his heels and toes almost simultaneously. In addition, when walking with slippers or the like, a frequency component unique to a human walking can be extracted from the vibration signal A, for example, a spectrum of a high frequency such as the frequency f4 of (3) may be generated. It becomes possible. Therefore, it is conceivable that the output signal of the A / D converter 3112a includes a result of a frequency having a strong spectrum. As will be described later, it is possible to distinguish vibrations caused by human walking from other vibrations based on the generation cycle of the first waveform 51a of the vibration signal A.
[0034]
FIG. 8 is a diagram showing the internal configuration of the information extracting means 12a.
[0035]
The information extraction unit 12a includes an action information extraction unit 121a, a walking mode information storage unit 122a, an ID assignment unit 123a, and an ID storage unit 124a. The behavior information extraction unit 121a receives a signal of a detection result regarding vibration from the vibration detection unit 11a, and inputs walking state information (walking state information corresponding to the subject (for example, the aforementioned walking state information) stored in advance in the walking state information storage unit 122a. Then, only the walking information (vibration associated with walking) of the subject is output by extracting the frequency component of the vibration of the subject and the later-described vibration cycle)) and collating them. The ID assigning unit 123a receives the walking information, extracts an ID (an identification signal or other information for identification, which employs a number in this embodiment) stored in the ID storage unit 124a in advance, and obtains the walking information. And the walking information with the ID is output to the transmission means 13a. Note that the ID storage unit 124a is configured by a nonvolatile memory or the like, and is written at the time of shipment or installation.
[0036]
FIG. 9 shows a timing chart of a signal input to the behavior information extraction unit 121a.
[0037]
FIG. 9A shows a timing chart based on vibration generated when an object falls in a living space. Since the object has fallen, it is output only once. FIG. 9B shows a timing chart based on vibration when a human walks in a living space. Output is 5 times and cycle T ₁ It is continuous. Then, by comparing the signal of the detection result regarding the vibration from the vibration detecting means 11a with the walking mode information of the walking mode information storage unit 122a, the signal associated with the human motion (for example, walking) and the other signals are compared. It can be easily distinguished. For example, the period T is stored in the walking mode information storage unit 122a. ₁ Assuming that the timings of vibrations before and after and for three or more consecutive vibrations are stored, (2) in FIG. 9 can be determined to be human walking. In addition, it is conceivable to avoid the erroneous determination by adding a process such as weighting or averaging to the comparison determination of the vibration period. Further, information on the frequency component of the vibration is also stored in the walking mode information storage unit 122a, and by comparing the information with the result of the frequency component of the vibration included in the vibration signal from the vibration detecting unit 11a, it is more reliably determined whether or not the human is walking. In addition to being able to do this, it is possible to specify a person who is walking based on the habit of walking for each individual. In addition, since the determination is made based on the period of vibration and the frequency component of vibration, it is possible to distinguish the pet from a four-legged walking pet, and it is not necessary to consider the installation location unlike a conventional sensor using light.
[0038]
Next, the operation of the behavior mode extracting unit 1 will be described.
[0039]
FIG. 10 is a flowchart illustrating the operation of the behavior mode extracting unit 1. It is desirable that the walking mode information in the walking mode information storage unit 122a be obtained by having the subject actually walk at the time of installation and storing the data. The vibration detecting means 11 of the behavior mode extracting unit 1 detects the vibration (step [hereinafter abbreviated as ST] 1), and transmits a signal processing result of the vibration signal to the information extracting means 12 (ST2). In the process of ST3, the information extraction means 12 of the behavior mode extraction unit 1 collates the vibration signal with the previously stored walking mode information (related to the subject) (ST3). The collation content is the period of the vibration signal and the frequency component of the vibration as described above. When it is determined that the two do not match by the collation (ST4), it is regarded as noise such as vibration due to a fall of an object, and the apparatus enters a standby state for vibration detection (ST1). Therefore, it is desirable to perform collation based on the vibration information of a plurality of times. The behavior information extraction unit 121a of the behavior mode extraction unit 1 extracts the walking information of the subject (ST5), and the ID assignment unit 123a of the behavior mode extraction unit 1 assigns an ID to the walking information (ST6). In the process of ST7, the transmission unit 13 transmits (transmits) the walking information and the ID to the abnormality determination unit 2. Then, the process returns to ST1. Note that the collation by the behavior mode extraction unit 1 may be performed until the target person is specified, or only the target is determined by the abnormality determination unit 22 by determining whether or not the person is walking. You may.
[0040]
Next, an example of installation of the above-described behavior mode extraction unit 1 and abnormality determination unit 2 will be described.
[0041]
FIG. 11 shows an installation example of the behavior mode extraction unit 1 and the abnormality determination unit 2 in a living space 40 such as a house. In FIG. 11, five behavior mode extraction units 1a to 1e (n = 5) are shown as the behavior mode extraction unit 1. The living space 40 is partitioned into a living room 41, a Japanese-style room 42, a dressing room 43, a toilet 44, a kitchen 45, and the like, and the behavior mode extraction units 1a to 1e are installed at appropriate positions in each partition. In addition, the abnormality determination unit 2 is installed at a place suitable for receiving signals from the behavior mode extraction units 1a to 1e arranged at various places in the house 40 by wire or wirelessly.
[0042]
FIG. 12 shows an internal configuration of the abnormality determination unit 22 of the abnormality determination unit 2.
[0043]
The abnormality determination unit 22 includes a determination criterion storage unit 222 in which a determination criterion (for example, a threshold of the number of times of walking over a predetermined period) is stored, and a reception unit 21 transmitted from the transmission unit 13 of the behavior mode extraction unit 1 as described above. And an abnormality judging unit 221 for judging an abnormal state from the walking information received in the step S1 and the judgment standard. The abnormality determining unit 221 includes a timer 223 that counts time and time. The abnormality determining unit 221 compares the walking information with the determination criterion, and generates an abnormal signal when it is determined from the counted time that it is abnormal, and sends the abnormal signal to the communication unit 23. The criterion is determined by a predetermined report destination such as a call center.
[0044]
Next, the operation of the abnormality determination unit 2 will be described.
[0045]
FIG. 13 is a flowchart illustrating the operation of the abnormality determination unit 2. The reception means 21 of the abnormality determination unit 2 is in a standby state for receiving the walking information and the ID described above (ST10). When the walking information and the ID are received, the abnormality determining unit 221 extracts the walking information and the ID received by the receiving unit 21 (ST11). Thereafter, the walking information and the criterion are collated to determine whether there is an abnormality (ST13). This determination is made, for example, when abnormal walking information such as the subject walking while dragging his / her foot is input. If this determination is "NO", the counter of the timer 223 is cleared, and the routine proceeds to ST10. On the other hand, while the signal is not received, the counter of the timer 223 is counted (ST12), and the data is constantly collated with the criterion to determine whether or not there is an abnormality (ST13). Specifically, as a criterion, it is conceivable that an abnormality is determined when the threshold value of the number of vibrations over a predetermined period cannot be cleared. For example, after the subject's walking vibration is detected by the behavior mode extracting unit 1 installed in the dressing room 43 (ST11), the subject's walking is again performed for 2 hours or more from the clothing room 43 or the behavior mode extracting unit 1 in another place. If no information is detected, it is determined that something has happened while the subject is taking a bath. In another example, if there is a large vibration after the walking information of the subject and there is no vibration information for a predetermined period thereafter, it is determined that the subject has fallen during walking. And so on. If it is determined to be abnormal, the process proceeds to ST14.
[0046]
In the process of ST14, abnormality determination section 221 generates an abnormality signal. Then, the communication means 23 transmits the abnormal signal to the outside (ST15), and proceeds to the processing of ST10. The outside refers to a specialized company that operates 24 hours a day, such as a call center, a family living away from home, or a mobile phone of a family going out.
[0047]
As described above, in the first embodiment of the present invention, information relating to human motion, particularly vibration generated by walking, is extracted from the output of the vibration detecting means 11, so that erroneous recognition and installation of the system as in the conventional system are performed. It is possible to eliminate trouble. In addition, by storing walking information (frequency components and vibration periods) of a plurality of subjects in advance, it is possible to separate and extract walking information for each individual, and as in the past, only a single subject alone was able to Instead, it is possible to provide the system to elderly people living with their families and the like, and a great effect can be obtained in practical use. In the case of a single person, it goes without saying that only the person walking may be extracted without specifying the target person from the walking information.
[0048]
[Second embodiment]
FIG. 14 is a diagram illustrating the configuration of the life watching apparatus according to the second embodiment, and illustrates the configuration of the behavior mode extraction unit 1, the abnormality determination unit 2, and the network 3. The n behavior mode extraction units 1a to 1k each include vibration detection units 11a to 11k, information extraction units 12a to 12k connected downstream of the vibration detection units 11a to 11k, and transmission units 13a to 13k. Have been. Note that the configuration of the behavior mode extraction unit 1 and the installation position of the life watching device in the living space 40 are the same as those in the first embodiment, and a description thereof will be omitted. The abnormality determination unit 2 is a reception unit 21 connected to the transmission unit 13 of the behavior mode extraction unit 1 by wireless or wired, an analysis unit 24 that calculates the amount of activity from walking information, and determines an abnormal state of the subject. And a communication means 23 for notifying (transmitting) an abnormal state to the outside when the state is determined to be abnormal. The communication means 23 of the abnormality determination unit 2 is connected to a network (for example, the Internet or a telephone line) 3.
[0049]
FIG. 15 is a diagram showing the internal configuration of the analysis means 24.
[0050]
The analysis unit 24 of the second embodiment includes an analysis unit 241, an analysis reference storage unit 242 in which analysis standards (see FIG. 16 described later) are stored in advance, and an analysis unit that stores analysis results (see FIG. 17 described later). And a result storage unit 243. Further, the analysis unit 241 includes a clock unit 1241 for measuring time. The walking information transmitted from the receiving unit 21 is input to the analysis unit 241. The analysis unit 241 takes out the analysis reference stored in the analysis reference storage unit 242, calculates an “activity amount” described later, and stores the analysis result. Stored in the section 243. The detailed calculation of the “activity amount” will be described with reference to FIGS.
[0051]
FIG. 16 is a diagram showing the contents of the analysis criterion stored in the analysis criterion storage unit 242.
[0052]
FIG. 16 shows an example of an analysis reference table as an analysis reference. In the analysis reference table, the horizontal axis indicates the movement distance (of the subject) in increments of 1 to 5. When the moving distance is 5, the moving distance becomes the longest, and as the number decreases, the moving distance decreases, and when the moving distance is 1, the moving distance becomes the shortest. The vertical axis indicates the travel time (of the subject) in increments of 1 to 5. When the moving time is 5, the shortest moving time is obtained, and as the number decreases, the moving time becomes longer. When the moving time is 1, the longest moving time becomes. Since the moving distance and the moving time are determined by the installation location of the behavior mode extracting unit 1, it is desirable to set the moving distance and the moving time for each subject's home. The setting may be made at the time of installation, or may be calculated from the layout of the home and stored at the time of shipment.
[0053]
An activity amount (also referred to as a weight coefficient) is set in advance from the above-described moving distance and moving time. For example, when the moving distance is 5 and the moving time is 5 that is the shortest, the activity amount is 9, and when the moving distance is 1 and the moving time is the longest 1, The amount of activity is 1. The analysis reference table indicates the walking speed of the subject (the period T in FIG. 9). ₁ ) May be set, but is not limited to this.
[0054]
FIG. 17 is a diagram illustrating the contents of the analysis result stored in the analysis result storage unit 243.
[0055]
FIG. 17 shows an analysis result table as the analysis result. The current time is shown on the left side of the analysis result table, and the activity amount at that time is shown on the right side. For example, the activity is 6 at 10:05 on October 3, 2002, the activity is 0 at 10:10 on October 3, 2002, and the activity is being analyzed at 10:15. I have. In the analysis result table, the sum of the activity amounts at predetermined time intervals (every several minutes) may be input as shown in FIG. 17, or the signal detected by the behavior mode extraction unit 1 may be received by the receiving unit 21 and analyzed. It may be input every time the analysis is performed by the means 24.
[0056]
FIG. 18 is a graph of the analysis result table.
[0057]
On October 3, the activity of the resident 50 is recognized from about 6:00 to about 21:00, and a remarkable amount of activity of the resident 50 is recognized from about 9:00 to about 13:00. On the other hand, on October 4, the activity of the resident 50 is recognized from about 6:00 to about 21:00, but no remarkable activity amount of the resident 50 is recognized. Therefore, on October 3, the resident can determine that he has actively moved around in the living space 40, and on October 4, he can judge that he has not actively moved around in the living space 40. The analyzing unit 241 analyzes that the physical condition of the subject is not good due to a decrease in the activity amount, and notifies the abnormality determining unit 22 of the abnormality determining unit 2 of the fact. An abnormality can be notified to the outside together with the analysis result.
[0058]
Note that the result of the analysis unit 241 may be periodically transmitted to an external predetermined report destination via the communication unit 23 via the communication unit 23, not only when the result of the analysis unit 241 is determined to be abnormal. Thereby, not only the abnormality determination unit 2 but also a doctor or a nurse resident in a call center or the like can determine the abnormality of the target person, and the abnormality can be detected more quickly and reliably. Further, it is conceivable that the detection range of the vibration detecting means 11 is widened, the activity amount is calculated from the number of times of walking in the same room, and the calculated activity amount is added to the above-mentioned activity amount due to movement between rooms.
[0059]
[Third embodiment]
In the first and second embodiments, the subject lives in the living space 40, but in the third embodiment, it is assumed that the living space is replaced by a ward or a facility for the elderly.
[0060]
FIG. 19 is a diagram illustrating a configuration of a life watching device according to the third embodiment, and illustrates a configuration including n (n is an integer of 1 or more) behavior mode extraction units 1 and an abnormality determination unit 2. The behavior mode extracting unit 1 has the same configuration as that of the first and second embodiments, and a description thereof will be omitted. The abnormality determination unit 2 is configured to include a reception unit 21 connected to the transmission unit 13 of the behavior mode extraction unit 1 wirelessly or by wire, and an abnormality determination unit 22 for determining an abnormal state of the subject. Note that the abnormality determination unit 22 may be hardware, but may be configured by dedicated software in a personal computer (PC). Hereinafter, the case of software will be described.
[0061]
FIG. 20 is a diagram illustrating the internal configuration of the abnormality determination unit 22 according to the third embodiment. Note that the abnormality determination unit 22 is assumed to be a PC installed in a place such as a nurse center or a facility management room where people who care and care are gathered.
[0062]
The abnormality determination unit 22 includes an abnormality determination unit 221 and a criterion storage unit 222. The criterion for the determination by the abnormality determination unit 221 is input to the criterion storage unit 222 in advance by the criterion input unit 224. The information input in the determination criterion input unit 224 includes, for example, the subject name and the walking information of the subject, the ID number of each behavior mode extraction unit 1 and its installation location, and a method of responding to the subject. is there. The walking information and the ID are input from the receiving unit 21 to the abnormality determining unit 221, and the abnormality determining unit 221 determines an abnormality by comparing the walking information and the ID stored in the determination standard storage unit 222. When it is determined that there is an abnormality, the subject name and its contents are transmitted to the notifying means 70. The notification means 70 is a screen of a PC or the like, and notifies a staff member at the nurse center by pop-ups or beeps on the screen.
[0063]
FIG. 21 is a diagram showing an ID number and installation location information of each behavior mode extraction unit 1.
[0064]
FIG. 21 shows an installation location information table. On the left side, the ID number (identification number) of each behavior mode extracting unit 1 is shown, and there are numbers 1, 2, 3,... On the right side, the first room, the second room, the third room,..., The corridor, the entrance,.
[0065]
For example, when the subject is an elderly person with dementia, he or she may go out without the permission of facility staff. Therefore, the behavior mode extraction unit 1 storing the ID number 7 is installed at the entrance of the facility. The criterion storage unit 222 stores in advance criterion information that informs the staff when the subject comes to the entrance. When the subject approaches the entrance, the behavior mode extraction unit 1 extracts the walking information and transmits the information together with the ID number to the abnormality determination unit 22 of the nurse center wirelessly or by wire. Note that the behavior mode extraction unit 1 may determine only whether or not the person is walking, and the abnormality determination unit 22 may specify the target person from the walking information (frequency component or vibration cycle). In this case, it is not necessary for the behavior mode extraction unit 1 to wait for a large amount of walking information including a large number of subjects, so that the behavior mode extraction unit 1 can be made inexpensive. The abnormality determination unit 22 determines that the target person is at the entrance from the walking information and the ID number, and the notification unit 70 notifies (notifies) the staff at the nurse center. Upon receiving the report, the staff rushes to the entrance or locks the entrance key. Further, the door of the entrance may be controlled by an electronic key, and the key of the entrance door may be automatically locked based on the result of the abnormality determination.
[0066]
Further implementation modes at the facility are described below. FIG. 22 shows an installation example of the behavior mode extraction unit 1 and the abnormality determination unit 2 in the ward 60 (one floor in this case). In FIG. 22, twelve behavior mode extraction units 63a to 63j, 63x, 63y (n = 12) are shown as the behavior mode extraction unit 1. The ward 60 is divided into first to ninth rooms 61a to 61i (beds 62a to 62i are installed in each room), a nurse center 61j, a corridor 61x, an entrance 61y, and the like, and act at an appropriate position in each division. Aspect extraction units 63a to 63j, 63x, 63y are provided. Further, the abnormality determination unit 2 that receives signals from the behavior mode extraction units 63a to 63j, 63x, and 63y arranged in various places of the ward 60 by wire or wireless is installed in the nurse center 61j.
[0067]
In hospitals, etc., falling of the elderly is a problem. In particular, a patient who is unable to walk has a risk of getting tangled and falling when getting up from the bed and starting walking after getting up. Therefore, there has been proposed a sensor for detecting a leaving state in which an elderly person sleeping on a bed is going to get up. For example, a pressure sensor (bed sensor) is installed between the user and the bedding, and biological information such as body movement, breathing, and heart rate is measured to detect that the user has left the bedding. However, the output of a mat-type sensor or the like installed on the bedding differs depending on the state of the subject sleeping, for example, the subject may be turned sideways, or may react when slightly raised. Therefore, there is a problem in that, despite being in the bed, the person is erroneously determined to be out of bed and calls a staff member (caregiver) in the facility. The present embodiment solves such a problem, and it is possible to reliably detect that the device has been raised.
[0068]
FIG. 23 shows a lifestyle of the resident (patient) 50 in the first room 61a of the ward 60.
[0069]
The behavior mode extraction unit 63a may be located beside the bed or in a space under the bed, and may be anywhere as long as the vibration generated by the patient 50 during walking can be detected. In FIG. 23, the floor 30 vibrates as the patient 50 goes out of the bed 62a and walks in the room, and the behavior mode extracting unit 63a (the vibration detecting unit 11a) detects the vibration. When the patient 50 gets up from the bed and tries to walk out, the patient 50 gets up on the bed and prepares to leave the bed, and the vibration generated at that time may be detected as a prediction signal. The vibration detection unit 11a of the behavior mode extraction unit 63a extracts from the vibration information that the patient 50 has started walking away from the bed, adds an ID thereto, and transmits the information by wireless transmission by the transmission unit 13a.
[0070]
The abnormality determination unit 2 of the nurse center 61j receives a signal from the behavior mode extraction unit 63a. The abnormality determination unit 221 of the abnormality determination unit 2 performs processing in accordance with the criterion in the criterion storage unit 222. For example, by registering that the patient 50 is elderly and needs assistance for movement in the determination criterion storage unit 222, it is recognized from the received ID that the behavior mode extraction unit 63a is near the patient 50. Then, the notification unit 70 notifies the staff. In addition, the walking information of the patient 50 is stored in advance in the criterion storage unit 222, and the transmitted vibration information is compared with the walking information of the patient 50, whereby the patient 50 can be more reliably recognized. Since the notification is made when the patient 50 gets up and starts walking, it is possible to avoid a false report due to the operation on the bed as in the related art, and it is possible to reduce the burden of rushing for the staff.
[0071]
FIG. 24 is a diagram illustrating the operation of the abnormality determination unit 2.
[0072]
In the process of ST20, the walking information of the patient 50 and the corresponding method are input in advance in the criterion input unit 224. The response method is, for example, that when the behavior information extraction unit 63a receives the walking information of the patient 50, the fact is notified and the staff rushes. When the patient 50 starts walking, the receiving unit receives the walking information and the ID transmitted from the behavior mode extracting unit 63a (ST21), and extracts the information (ST22). The criterion is read from the criterion storage unit 222 based on the received walking information and ID, and collated (ST23). As a result of the collation (ST24), it is determined that the patient is the patient 50, and if a measure is necessary, the measure is notified (ST25). On the other hand, when the staff or the like enters the patient's room 61a for nursing, the stored walking information does not match, so that the process returns to the standby for reception (ST21) without taking any action such as notification.
[0073]
In the third embodiment, as in the second embodiment, a plurality of behavior mode extraction units can be arranged in a facility to collect the amount of activity for each patient. Needless to say, the abnormality determination unit 2 can be connected to a hospital network or a public line. Furthermore, if the device is installed in a place where a fall accident is likely to occur, such as a toilet, it is possible to grasp who entered from the walking information and detect the fall by vibration, so that the fall of the patient can be detected quickly. It is also possible.
[0074]
The embodiment has been described above, but the present invention is not limited to this. For example, as a vibration sensor for detecting vibration, an acceleration sensor, a speedometer, a microphone, a pressure gauge, a strain gauge, or the like may be used as appropriate.
[0075]
In addition, the abnormality determination means is incorporated in the behavior mode extraction unit, and an abnormal command is generated and transmitted wirelessly only when there is an abnormality, and the communication means receiving the abnormal command generates an abnormal signal and transmits it to the outside. Is also good. This makes it possible to minimize the power for wireless transmission when the behavior mode extraction unit is running on a battery.
[0076]
In the embodiment, a predetermined habit of vibration (specifically, a cycle of vibration) associated with walking of a resident is detected, and walking information of only the resident is extracted to determine an abnormal state of the resident. However, the present invention is not limited to this. For example, instead of the habit (period) of the vibration caused by the resident's walking, the resident who is not the target wears slippers and socks, and the resident who is the target uses the bare feet to Authentication may be performed to determine the abnormal state of the subject. If there is an elderly person (target person) who has a wandering habit among family members living together, if the person other than the target person wears slippers, it is easy to determine the walking information. Therefore, if the behavior mode extracting unit is installed at the entrance or the entrance, it is possible to easily construct a wandering prevention system.
[0077]
Further, since the vibration accompanying the occupant propagates on the floor, the vibration detecting means can be arranged at a corner of the room in the living space. Since it can be arranged at this position, the interior of the living space can be maintained widely.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of a life watching device of the present invention.
FIG. 2 is a diagram showing the configuration of a first embodiment of the life watching apparatus of the present invention.
FIG. 3 is a diagram showing a configuration in which there are a plurality of behavior mode extraction units.
FIG. 4 is a diagram illustrating a configuration of a vibration detection unit.
FIG. 5 is a diagram illustrating generation of vibration.
FIG. 6 is a diagram showing vibration signals at points A and B in FIG. 4;
FIG. 7 is a diagram illustrating frequency components of vibration.
FIG. 8 is a diagram showing a configuration of information extraction means.
FIG. 9 is a diagram showing a noise cut in the behavior information extraction unit.
FIG. 10 is a flowchart showing the operation of the behavior mode extraction unit.
FIG. 11 is a diagram showing an installation example of the watching device of the present invention in a living space.
FIG. 12 is a diagram illustrating a configuration of an abnormality determination unit.
FIG. 13 is a flowchart illustrating the operation of the abnormality determination unit.
FIG. 14 is a diagram showing a configuration of a life watching device according to a second embodiment.
FIG. 15 is a diagram illustrating a configuration of an analysis unit according to the second embodiment.
FIG. 16 is a diagram showing an analysis reference table.
FIG. 17 is a diagram showing an analysis result table.
FIG. 18 is a diagram illustrating a relationship between time and an activity amount.
FIG. 19 is a diagram showing a configuration of a life watching device according to a third embodiment.
FIG. 20 is a diagram illustrating a configuration of an analysis unit and a configuration of a notification unit according to a third embodiment;
FIG. 21 is a diagram showing an installation location information table.
FIG. 22 is a diagram showing an installation example of the watching device of the present invention in a ward.
FIG. 23 is a diagram showing a mode of detecting vibration accompanying a patient walking in a ward.
FIG. 24 is a flowchart showing the operation of the abnormality determination unit.
[Explanation of symbols]
1, 1a to 1k, 63a to 63j, 63x, 63y ... behavior mode extraction unit, 2 ... abnormality determination unit, 3 ... network, 11 ... vibration detection unit, 12 ... information extraction unit, 13 ... transmission unit, 21 ... reception unit , 22 ... abnormality judgment means, 23 ... communication means, 24 ... analysis means, 30 ... floor, 40 ... living space, 41 ... living room, 42 ... Japanese room, 43 ... dressing room, 44 ... toilet, 45 ... kitchen, 50 ... living Person, 50a ... right foot, 50b ... left foot, 51 ... vibration signal, 51a ... first waveform, 51b ... second waveform, 60 ... ward, 61a-61i ... room, 61j ... nurse center, 61x ... hallway, 61y ... entrance, 62a to 62i: bed, 70: notification means, 111a: vibration sensor, 112a: signal processing unit, 121a: action information extraction unit, 122a: walking mode information storage unit, 123a: ID assignment unit, 124a ... D storage unit, 221: abnormality determination unit, 222: determination criterion storage unit, 223, 1241: clock unit, 224: determination criterion input unit, 241: analysis unit, 242: analysis criterion storage unit, 243: analysis result storage unit, 244 installation location information storage unit, 1112a amplifier, 2112a filter, 3112a A / D converter.

Claims (5)

Vibration detecting means installed in the living space where the subject lives, and detecting vibration generated in the living space;
Information extraction means for extracting information about vibration generated by human motion in the living space from the output of the vibration detection means,
And a transmission unit for transmitting the information extracted by the information extraction unit. 2. The life watching device according to claim 1, wherein the walking of the subject is specified from the information on the vibration. The life monitoring apparatus according to claim 1 or 2, further comprising: an abnormality determining unit configured to determine the abnormal state of the subject by comparing the extracted information with a predetermined determination criterion;
A communication means for notifying a result of the abnormality determination means to a predetermined notification destination specified in advance; The life watching device according to any one of claims 1 to 3,
A plurality of each are installed in the living space of the target person,
An ID assigning unit that assigns an ID corresponding to the information extracting unit to the information extracted by each of the plurality of information extracting units,
The life watching apparatus, wherein the transmission unit transmits the ID together with the extracted information. The life watching device according to claim 4,
A life watching apparatus, comprising: an analyzing unit that calculates an activity amount of the target person from information extracted from a living space of the target person.

Images