JP6984309B2 - Behavior judgment system, behavior judgment method, and behavior judgment program - Google Patents

Description

The present invention relates to a behavior determination system, a behavior determination method, and an action determination program.

In recent years, by analyzing sensor data (which may be referred to as measurement data) acquired by a wearable device (which may be referred to as a wearable sensor) equipped with various sensors such as an accelerometer, human behavior can be detected. There are efforts to provide judgment technology (which may be referred to as behavior sensing technology).

For example, in the prior art, human behavior is determined by calculating a feature amount of human behavior based on sensor data acquired from a wearable device and comparing it with a pre-registered reference pattern (Patent Document 1, etc.). reference).

Japanese Unexamined Patent Publication No. 2009-134590 Japanese Unexamined Patent Publication No. 10-113343 Japanese Unexamined Patent Publication No. 2008-102951

In the conventional technology, human behavior that can be analyzed from sensor data acquired from a wearable device is only a simple movement such as "walking", "running", "standing up", and "sitting", and has a high meaning of human behavior. It is difficult to find.

As an attempt to improve this, we used a classifier to identify behaviors according to the subject's current position, assuming that the work behaviors that the subject can take are limited according to the work location. There is also a prior art that proposes to more accurately identify the complex behavior of the subject. According to this technology, when the work place is, for example, an "ophthalmology examination room", complicated behaviors such as "walking / moving", "document work", "ophthalmology examination assistance", "hygiene treatment", and "patient transfer" are performed. Can be identified.

However, in order to determine the advanced meaning of the subject's behavior, the more complicated the behavior to be identified, the larger the amount of reference patterns to be learned in advance, and the more the calculation cost is wasted. Even if the actions that can be taken are limited depending on the work place, if there are a plurality of actions that can be taken, the calculation cost is further wasted.

It is an object of the present disclosure to provide a behavior determination system, a behavior determination method, and a behavior determination program capable of determining the advanced meaning of the behavior of a subject wearing a wearable device.

According to one aspect of the disclosure, the behavioral determination system captures measurement data from a sensor mounted on the subject, an acquisition means, and a unit movement, which is an individual movement constituting the behavior that the subject can take. Based on the unit feature amount for identification, the first determination means for determining the unit operation from the measurement data and the first determination result indicating the unit operation determined by the first determination means are first determined. Registered in the result table, the valid time of the first judgment result registered in the first judgment result table is managed, and the first judgment result after the valid time is deleted from the first judgment result table. The action that the subject can take is determined based on the combination of the effective time management means and one or more of the unit movements indicated by the one or more valid first determination results registered in the first determination result table. The second determination means and the adjustment means for adjusting the effective time in the effective time management means are provided , and the adjustment means is excluded from the combination of the unit operations in the determination by the second determination means. When the first operation is detected based on the measurement data, the first adjustment means for adjusting the effective time in the effective time management means to be extended, and the first determination result in the first determination result table. It has at least one of a second adjusting means for adjusting so that the effective time in the effective time management means is shortened when the moving distance of the subject after the registration is equal to or greater than the threshold value. To.

According to one aspect of the disclosure, the higher meaning of the subject's behavior can be determined.

FIG. 1 is a diagram showing an example of the configuration of the behavior determination system according to the first embodiment. FIG. 2 is a diagram showing a content example of the first pattern DB (T110) according to the first embodiment. FIG. 3 is a diagram showing a content example of the second pattern DB (T120) according to the first embodiment. FIG. 4 is a diagram showing an example of a processing flow in the behavior determination system 10 according to the first embodiment. FIG. 5 is a diagram showing an example of the configuration of the behavior determination system 10 according to the second embodiment. FIG. 6 is a diagram showing a content example of the first determination result table T130A according to the second embodiment. FIG. 7 is a diagram showing an example of a processing flow in the behavior determination system 10 according to the second embodiment. FIG. 8 is a diagram showing an example of the flow of the effective time management process according to the second embodiment. FIG. 9 is a diagram showing an example of the configuration of the behavior determination system 10 according to the third embodiment. FIG. 10 is a diagram showing an example of the flow of the effective time management process (S104A) according to the third embodiment. FIG. 11 is a diagram showing a content example of the first pattern DB (T110) according to the third embodiment. FIG. 12 is a diagram showing an example of the configuration of the behavior determination system 10 according to the fourth embodiment. FIG. 13 is a diagram (No. 1) showing an example of the flow of the effective time management process (S104A) according to the fourth embodiment. FIG. 14 is a diagram (No. 2) showing an example of the flow of the effective time management process (S104A) according to the fourth embodiment. FIG. 15 is a diagram showing an example of the configuration of the behavior determination system 10 according to the fifth embodiment. FIG. 16 is a diagram showing a content example of the action setting table T140D according to the fifth embodiment. FIG. 17 is a diagram showing an example of a processing flow in the behavior determination system 10 according to the fifth embodiment. FIG. 18 is a diagram showing a content example of the event log T150D according to the fifth embodiment.

The inventors of the present invention have conducted diligent studies in the field of behavior sensing technology for determining the behavior of a human being (subject, may be referred to as a subject) by analyzing sensor data. As a result, it was concluded that it is difficult to determine a sufficiently high degree of meaning from the sensor data acquired by the wearable device in the prior art.

For example, the more complicated the behavior to be identified, the larger the amount of reference patterns to be learned in advance, and the more the calculation cost is wasted. Moreover, if there are a plurality of actions to be identified, the calculation cost is further wasted.

The inventors of the present invention have found that even if the subject's behavioral complexity increases, it can be represented by a combination of relatively simple movements. For example, the action of "locking the front door" is a series of relatively simple actions such as "opening the door", "walking / moving", "inserting the key", "twisting the key", and "unlocking the key". It can be expressed by a combination of various movements (which may be called a unit movement).

In addition, the inventors of the present invention have noticed that relatively simple movements can be duplicated in a plurality of actions by focusing on individual movements included in a complicated action. For example, the action of "using the door" is "opening the door", "walking / moving", "closing the door", "twisting the key", "sitting", "standing up", "twisting the key", " It can be expressed by a combination of a series of unit movements such as "opening the door", "walking / moving", and "closing the door". In this example, the action of "using the toilet" overlaps with the action of "locking the front door" in some unit actions. In this case, if the reference pattern (which may be called a feature amount) is learned for each action to be identified as in the prior art, the learning amount for the overlapping unit movements becomes redundant and the calculation cost is wasted. Will be done.

According to one aspect of the disclosure below, it is based on a combination (may be referred to as a second pattern) of reference patterns (unit patterns, unit features, may be referred to as a first pattern) that define unit operation. Therefore, a technique capable of determining a high degree of meaning of a subject's behavior is provided.

Hereinafter, embodiments for carrying out the present invention (hereinafter, may be referred to as embodiments or examples) will be described with reference to the drawings. The configuration of the embodiment shown below is an example for embodying the technical idea of the present invention, and is not intended to strictly limit the present invention to the configuration of the following embodiment. Equally applicable to other embodiments included in the claims. For example, the subject wearing the wearable device may be a human, a pet animal (dog, cat, etc.), a domestic animal (cow, horse, sheep, etc.), or a wild animal (tiger, bear, migratory bird, etc.). It may be a non-human animal such as, or it may be a movable device such as a robot. In the following disclosure, for the sake of brevity of explanation, the explanation assuming a human being may be used as an example of a subject. However, the behavior (reference pattern) to be identified may be appropriately changed according to the attributes of the subject (biological classification of humans, dogs, cats, cows, horses, sheep, etc., type of device, etc.). Note that it is not intended to limit the body to humans. The subject in the present disclosure is an object whose behavior is determined by the behavior determination system disclosed below, and may be referred to as an action subject, for example. However, it should be noted that even the movement of the subject with the assistance of a third party can be referred to as the movement subject of the movement.

<Example 1> In the action determination system 10 according to the first embodiment, a first determination based on a unit feature amount (which may be referred to as a first pattern) for identifying a unit operation and one or more unit operations are performed. A second determination based on a combination (which may also be referred to as a second pattern) determines the high degree of meaning of the subject's behavior.

FIG. 1 is a diagram showing an example of the configuration of the behavior determination system 10 according to the first embodiment. The behavior determination system 10 shown in FIG. 1 includes a processor 110, a memory 120, and a motion sensor 130. Each of the processor 110, the memory 120, and the motion sensor 130 may be communicably connected wirelessly or by wire. For example, the processor 110 may be configured to be able to read and write programs and data stored in the memory 120. For example, the processor 110 may be configured to be able to read the measurement data (information about the measured physical quantity, which may be referred to as sensor data) output from the motion sensor 130. Further, in the configuration of the behavior determination system 10 illustrated in FIG. 1, each component such as the processor 110, the memory 120, and the motion sensor 130 is illustrated, but this embodiment is limited to this. is not it. For example, the behavior determination system 10 may have one or more processors 110, one or more memories 120, and one or more motion sensors 130. Then, the internal configurations described later may be duplicated among the same components, or the internal configurations may be distributed and arranged so as not to be duplicated. Although not shown in FIG. 1, the behavior determination system 10 according to the first embodiment may include one or more power sources such as a rechargeable battery. It should be noted that in the present disclosure, the information regarding the measured physical quantity may be simply referred to as information, data, or signal.

The processor 110, for example, reads and executes a program (may be referred to as an action determination program) stored in the memory 120 to realize the processing according to the first embodiment (referred to as a computer). May be). In other words, the processor 110 has an aspect as an execution subject of the process according to the first embodiment. Examples of the processor 110 include a CPU (Central Processing Unit), an MPU (Micro Processing Unit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), and the like. The processor 110 may be a multi-core processor including two or more cores.

The processor 110 shown in FIG. 1 has a data acquisition unit P110, a first determination unit P120, and a second determination unit P130. These means (P110 to P130) may be realized, for example, by reading and executing a program stored in the memory 120 by the processor 110. In other words, the processor 110 realizes the functions of the data acquisition means P110, the first determination means P120, and the second determination means P130 by executing a predetermined program (which may be referred to as an action determination program). Converted to a hardware circuit. The action determination program may be stored in advance inside the processor 110.

The data acquisition means P110 is configured to acquire measurement data from the motion sensor 130 mounted on the subject.

The first determination means P120 is configured to determine the unit motion from the measurement data based on the unit feature amount for identifying the unit motion which is an individual motion constituting the action that the subject can take.

The second determination means P130 is configured to determine possible actions by the subject based on a combination of one or more unit movements obtained by performing the first determination a plurality of times.

The memory 120 is a circuit configured to store and hold data and programs (which may be referred to as an action determination program) related to various processes executed by the processor 110. The memory 120 includes at least one of a non-volatile storage device and / or a non-volatile storage device. For example, RAM (Random Access Memory), ROM (Read Only Memory), SSD (Solid State Drive), HDD (Hard Disk Drive) and the like can be mentioned. In FIG. 1, the memory 120 is a general term for various storage devices such as a main storage device and an auxiliary storage device. The action determination program may be stored in the memory 120 in advance, or is downloaded from a device (for example, an information processing device such as a smartphone or a server) connected via a communication circuit (not shown) and stored in the memory 120. You may.

When attached to the subject, the motion sensor 130 is configured to output sensor data (which may be referred to as measurement data) having a signal strength corresponding to the movement of the subject. For example, the motion sensor 130 may be an acceleration sensor that detects the acceleration of the subject. The accelerometer used for such an application may be a three-axis accelerometer capable of detecting acceleration in three directions of the X-axis, the Y-axis, and the Z-axis. For example, if the subject is a human, the motion sensor 130 may be worn on the subject by a wristband fastened to the subject's wrist. When the subject is a non-human animal, the motion sensor 130 may be attached to the subject by a band fastened to the subject's neck or the like. The part of the body on which the motion sensor 130 is attached is not limited to these, and may be attached to other parts or may be attached to a plurality of parts.

In FIG. 1, the memory 120 has a first pattern DB (T110) and a second pattern DB (T120).

The first pattern DB (T110) is information (first definition information) that defines a unit feature amount (may be referred to as a first pattern) for a unit movement, which is an individual movement constituting a behavior that a subject can take. It may be referred to as).

The second pattern DB (T120) may be referred to as information (may be referred to as second definition information) that defines a combination of one or more unit movements (may be referred to as a second pattern) with respect to the actions that the subject can take. ) Is stored.

FIG. 2 is a diagram showing a content example of the first pattern DB (T110) according to the first embodiment. In FIG. 2, the first pattern DB (T110) has first definition information (unit operation information) having a unit operation ID (T110-1), a unit operation label T110-2, and a first pattern T110-3. May be referred to).

The unit operation ID (T110-1) is information used to uniquely identify the first definition information, and may be described using numbers, characters, symbols, or a combination thereof. In the example of FIG. 2, a total of eight unit operation IDs (T110-1) of "A1" to "A8" are shown. In other words, FIG. 2 illustrates eight first definition information.

The unit operation label T110-2 is information describing the name of the unit operation indicated by the first definition information. In FIG. 2, for example, the unit operation label T110-2 "open the door" describes the unit operation in which the subject opens the door. The unit movement label T110-2 "close the door" describes the unit movement in which the subject closes the door. The unit movement label T110-2 “walking / moving” describes a unit movement in which the subject walks or moves. The unit operation label T110-2 "insert" describes a unit operation in which a subject inserts a key or the like into a keyhole or the like. The unit operation label T110-2 "pull out" describes a unit operation in which a subject pulls out a key or the like from a keyhole or the like. The unit movement label T110-2 "twist the wrist" describes the unit movement in which the subject twists the wrist. The unit movement label T110-2 "sitting" describes the unit movement of sitting, which is an exercise that changes the body position from the standing state (standing position) to the sitting state (sitting position) of the subject. The unit movement label T110-2 "stand up" describes the unit movement of standing up, which is an exercise in which the subject changes his / her body position from a sitting position to a standing position.

The first pattern T110-3 is information indicating a feature amount (may be referred to as a unit feature amount) for identifying a unit operation. For example, the frequency characteristic of the measurement data obtained from the motion sensor 130 mounted on the subject when the subject performs a unit operation in the learning stage may be set as the first pattern T110-3. In this case, the method of attaching the motion sensor 130 to the subject may be appropriately adjusted so that the frequency characteristics of each unit operation are different.

As a method of acquiring the frequency characteristic from the measurement data which is a time series signal, a known method may be used. For example, by using a Fourier transform (which may be called time-frequency analysis), which is a method of signal analysis, it is possible to acquire frequency characteristics indicating the intensity of each frequency component from a time series signal (measurement data). Can be done. Further, as the unit feature amount, a signal characteristic other than the frequency characteristic may be used. For example, a value based on the amplitude value of the measurement data (time series signal) acquired from the motion sensor 130 may be used.

FIG. 3 is a diagram showing a content example of the second pattern DB (T120) according to the first embodiment. In FIG. 3, the second pattern DB (T120) stores the second definition information having the action ID (T120-1), the action label T120-2, and the second pattern T120-3.

The action ID (T120-1) is information used to uniquely identify the second definition information, and may be described using numbers, letters, symbols, or a combination thereof. In the example of FIG. 3, a total of two action IDs (T120-1) of "B1" and "B2" are shown. In other words, FIG. 3 illustrates two second definition information.

The behavior label T120-2 is information describing the behavior of the subject identified by the combination of unit movements (which may be referred to as the second pattern) shown in the second definition information. In FIG. 3, for example, the action label T120-2 "locking the front door" describes the behavior of the subject in daily activities of closing the front door when going out. The behavior label T120-2 "use the toilet" describes the behavior of using the toilet in the subject's daily activities.

The second pattern T120-3 is information indicating a combination of one or more unit movements for the behavior of the subject described by the behavior label T120-2. In FIG. 3, in the second definition information of the action label T120-2 “locking the front door”, the unit action ID (T110-1) is “A1” and the unit action ID (opens the door) and the unit action ID. (T110-1) is the unit operation (walking / moving) of "A3", the unit operation ID (T110-1) is the unit operation (insertion) of "A4", and the unit operation ID (T110-1) is "A6". A second pattern T120-3 indicating a combination of the unit movement (twisting the wrist) and the unit movement (pulling) whose unit movement ID (T110-1) is “A5” is set. Further, in FIG. 3, in the second definition information of the action label T120-2 “use the toilet”, the unit action ID (T110-1) is “A1” and the unit action ID (open the door) and the unit action ID ( The unit operation (walking / moving) where T110-1) is "A3", the unit operation (close the door) where the unit operation ID (T110-1) is "A2", and the unit operation ID (T110-1) is " The unit movement of "A6" (twisting the wrist), the unit movement of the unit movement ID (T110-1) is "A7" (sitting), and the unit movement of the unit movement ID (T110-1) is "A8" (standing up). And the unit movement (twisting the wrist) whose unit movement ID (T110-1) is "A6" and the unit movement (opening the door) whose unit movement ID (T110-1) is "A1" are shown. Two patterns T120-3 are set.

FIG. 4 is a diagram showing an example of a processing flow in the behavior determination system 10 according to the first embodiment. The processing flow shown in FIG. 4 may be repeatedly executed, for example, at predetermined time intervals. Alternatively, the processing flow shown in FIG. 4 may start execution in response to the timing at which the measurement data is output from the motion sensor 130 at predetermined time intervals.

First, the processor 110 executes a data acquisition process (S101). In S101, the processor 110 acquires measurement data from the motion sensor 130 mounted on the subject. Any method may be used to acquire the measurement data. For example, the processor 110 may acquire the measurement data by accessing the motion sensor 130 and reading the measurement data stored in the internal memory included in the motion sensor 130. Alternatively, the processor 110 may acquire the measurement data by receiving the measurement data actively output from the motion sensor 130. At that time, the processor 110 may acquire the measurement data actively output from the motion sensor 130 via the memory 120.

Next, the processor 110 executes the first determination process (S102). In S102, the processor 110 is a unit from the measurement data based on a unit feature quantity (which may be referred to as a first pattern) for identifying a unit movement which is an individual movement constituting the behavior that the subject can take. Judge the operation. For example, the processor 110 may acquire the frequency characteristics of the measurement data by performing a Fourier transform on the measurement data acquired in S101 in the first determination process of S102.

In S102, the processor 110 may determine the unit operation by comparing the frequency characteristic of the measurement data with the first pattern T110-3 of the first definition information stored in the first pattern DB (T110). .. For example, the processor 110 selects the first definition information indicating the first pattern T110-3 having the highest degree of correlation among the first patterns T110-3 in which the degree of correlation with the frequency characteristic of the measurement data is equal to or greater than the threshold value. Then, the unit operation may be determined.

In S102, the processor 110 may store the determination result of the first determination process for a plurality of times in the memory 120 by storing and holding the unit operation which is the determination result of the first determination process in the memory 120.

Then, the processor 110 executes the second determination process (S103). In S103, the processor 110 determines the actions that the subject can take based on the combination of one or more unit actions obtained by executing the first determination process a plurality of times. For example, the processor 110 reads the determination results of the first determination process for a plurality of times stored in the memory 120 to acquire a combination of one or more unit operations obtained by executing the first determination process a plurality of times. You may.

In S103, the processor 110 has a determination result (which may be referred to as a first determination result) of the first determination process for a plurality of times and a second pattern of the second definition information stored in the second pattern DB (T120). The subject may be compared with T120-3 to determine possible actions. For example, the processor 110 has a sequence of unit operation IDs (T110-1) indicated by each of the determination results of the first determination process for a plurality of times (may be referred to as a sequence of the first determination results) and second definition information. The behavior of the subject may be determined by comparing with the sequence of unit motion IDs (which may be referred to as the sequence of the second pattern) indicated by the second pattern T120-3.

In S103, for example, the processor 110 sequentially orders from the unit operation ID (T110-1) located at the end in the order of the unit operation ID (T110-1) indicated by each of the determination results of the first determination process for a plurality of times. It may be compared with the order sequence of the unit operation ID shown by the second pattern T120-3 of the second definition information.

In S103, the processor 110 has the unit operation ID indicated by the second pattern T120-3 of the second definition information in the order of the unit operation ID (T110-1) indicated by each of the determination results of the first determination process for a plurality of times. When a pattern matching the sequence is detected, the second definition information may be selected as the determination result of the second determination process.

In the second determination process of S103, in addition to the above method, a known matching method for determining the degree of matching between the two sequences is used, and the second pattern T120-3 most similar to the sequence of the first determination result is used. The second definition information having the above may be selected.

In S103, the processor 110 selects the second definition information based on the order of the first determination result, so that the behavior of the subject composed of the combination of the plurality of unit movements can be determined.

The processor 110 may store the determination result of the second determination process in S103 in the memory 120 as an event log. For example, the processor 110 may store the second definition information, which is the determination result of the second determination process, in the event log together with the information regarding the date and time indicated by the system clock at the time when the determination result of the second determination process is obtained. good. As a result, the behavior of the subject and the date and time can be grasped by referring to the second definition information stored in the event log and the information regarding the date and time.

According to one aspect of the first embodiment disclosed above, it is possible to grasp only the simple movement of the subject only by the judgment result of the first judgment processing based on the feature amount of the unit movement which is a relatively simple movement. Although it cannot be done, further, by executing the second determination process based on the combination of unit movements, it is possible to determine a higher meaning of the subject's behavior.

According to another aspect of the first embodiment disclosed above, the first determination process based on the feature amount of the unit operation, which is a relatively simple operation, and the second determination process based on the combination of the unit operations are performed. By combining them, it is possible to reduce the redundancy in determining the complex behavior of the subject. Such an action is useful for suppressing the waste of calculation cost when determining the advanced meaning of the behavior of the subject wearing the wearable device, and for more efficiently using the hardware resource.

<Example 2> In the behavior determination system 10 according to the second embodiment, the time for accumulating the result of the first determination process (which may be referred to as the first determination result) is managed, and the effective time is determined after the accumulation. The elapsed first determination result is excluded from the determination target of the second determination process.

FIG. 5 is a diagram showing an example of the configuration of the behavior determination system 10 according to the second embodiment. The behavior determination system 10 according to the second embodiment shown in FIG. 5 is the behavior determination system 10 according to the first embodiment shown in FIG. 1 in that the effective time management means P140A and the first determination result table T130A are added. It is different from, and is the same in other respects. Therefore, in FIG. 5, the same reference numerals are given to the configurations similar to those in FIG. 1, and the description of the same configurations will be omitted for the sake of brevity.

The effective time management means P140A registers the first judgment result indicating the unit operation determined by the first judgment means P120 in the first judgment result table T130A, and one or more firsts registered in the first judgment result table T130A. It is configured to manage each valid time of the determination result. Then, the effective time management means P140A deletes the first determination result after the effective time has passed from the first determination result table T130A.

Here, the initial value of the effective time may be set based on, for example, the time required for a series of unit movements constituting the actions that the subject can take (may be referred to as a required time). For example, at the stage of setting the effective time, the elapsed time from the start of a series of unit movements constituting the possible actions of the subject to the end of all the series of unit movements was measured multiple times and averaged. The initial value of the effective time may be set by using a value obtained by adding a predetermined multiple (for example, 3 times) of the standard deviation σ to the value (which may be referred to as the average measurement time). This makes it possible to estimate the effective time based on the time normally assumed for the behavior that the subject can take. As an example of the effective time set in this way, it may be, for example, 1.5 minutes.

The first determination result table T130A is configured so that the first determination result indicating the unit operation determined by the first determination means P120 is stored together with the information regarding the date and time.

FIG. 6 is a diagram showing a content example of the first determination result table T130A according to the second embodiment. In FIG. 6, the first determination result table T130A stores the first determination result information having the first determination result T130-1 and the registration date and time T130-2.

The first determination result T130-1 has information regarding the first definition information indicating the unit operation determined by the first determination means P120. In the example of the first determination result T130-1 shown in FIG. 6, the unit operation ID (T110-1) is set in the first definition information. The first determination result T130-1 in this embodiment is not limited to this, and may have a unit operation ID (T110-1) exemplified in FIG. 2 and a unit operation label T110-2. ..

The registration date and time T130-2 has information regarding the date and time when the first determination result by the first determination means P120 is registered in the first determination result table T130A. In the example of the registration date and time T130-2 shown in FIG. 6, information indicating a year, a month, and a day and information indicating an hour, a minute, and a second are shown. For example, the registration date and time T130-2-1 indicates information indicating "October 5, 2017" and information indicating "8:06:00". For example, according to the registration date and time T130-2-1 in FIG. 6, the date and time when the first determination result T130-1 indicating the unit operation ID “A1” is registered in the first determination result table T130A is “October 5, 2017”. It can be understood that it is "8:06: 0".

FIG. 7 is a diagram showing an example of a processing flow in the behavior determination system 10 according to the second embodiment. The process flow according to the second embodiment shown in FIG. 7 is different from the process flow according to the first embodiment shown in FIG. 4 in that an effective time management process (S104A) is added, and is the same in other respects. Is. In the example of FIG. 7, the effective time management process (S104A) is located between the first determination process (S102) and the second determination process (S103), but the execution order of the effective time management process (S104A) is appropriate. You may change it. For example, the effective time management process (S104A) may be executed after the second determination process (S103) is executed. In the following, for the sake of brevity, the description of the effective time management process (S104A) will be mainly described, and the detailed description of the other processes will be omitted.

For example, the processor 110 executes the effective time management process (S104A) after executing the first determination process (S102), so that the first determination result indicating the unit operation determined by the first determination process (S102) is shown. Is registered in the first determination result table T130A. In S104A, the processor 110 manages the valid time of each of the one or more first judgment results registered in the first judgment result table T130A, and the first judgment result after the valid time has passed is obtained from the first judgment result table T130A. delete.

In the second determination process (S103), the processor 110 determines an action that the subject can take based on one or more valid first determination results registered in the first determination result table T130A. In S103, the processor 110 may determine that the first determination result whose elapsed time from being registered in the first determination result table T130A does not exceed the effective time is a valid first determination result. Alternatively, in S103, the processor 110 may determine that the first determination result that has not been deleted from the first determination result table T130A by the effective time management process (S104A) is a valid first determination result.

FIG. 8 is a diagram showing an example of the flow of the effective time management process (S104A) according to the second embodiment. For example, the processor 110 executes the effective time management process (S104A) after executing the first determination process (S102), so that the process flow shown in FIG. 8 is executed.

The processor 110 registers the first determination result indicating the unit operation determined by the first determination process (S102) in the first determination result table T130A (S104A-1). In S104A-1, the processor 110 registers the unit operation ID (T110-1) among the information elements of the first definition information shown in the first determination result in the first determination result table T130A together with the registration date and time. You may. Here, the information set having the first determination result and the registration date and time may be referred to as the first determination result information.

The processor 110 searches the first determination result information registered in the first determination result table T130A for the first determination result information whose elapsed time from the registration date and time exceeds the valid time (S104A-2).

Based on the search result of S104A-2, the processor 110 determines whether or not there is first determination result information that exceeds the valid time (S104A-3). For example, when one or more first determination result information is present in the search result of S104A-2, the processor 110 may determine that there is a first determination result that exceeds the valid time (YES in S104A-3). ). On the other hand, if there is no first determination result information in the search result of S104A-2, the processor 110 may determine that there is no first determination result exceeding the effective time (NO in S104A-3). ..

When the processor 110 determines that there is a first determination result that exceeds the valid time (YES in S104A-3), the processor 110 deletes the first determination result information that exceeds the valid time from the first determination result table (S104A-). 4).

On the other hand, when it is determined that there is no first determination result exceeding the effective time (NO in S104A-3), the processor 110 skips without executing S104A-4.

The above is an example of the flow of the effective time management process (S104A) according to the second embodiment.

Next, a specific example of the effective time management process (S104A) according to the second embodiment will be described with reference to the example of the first determination result table T130A shown in FIG. In the example of FIG. 6, the first determination indicating that the unit operation ID is "A1" (that is, "open the door") in association with the registration date and time T130-2-1 "2017/10 / 05 --08: 06: 00". The result T130-1 is stored. The first judgment result T130- indicating that the unit operation ID is "A3" (that is, "walking / moving") in association with the next registration date / time T130-2-2 "2017/10/05 --08: 06: 02". 1 is stored. These unit movements may be due to the subject opening the door of a room in the house and moving through the corridor. Alternatively, the subject may have opened the front door of the house and moved out of the front door.

In the example of FIG. 6, the unit operation ID indicates "A1" (that is, "open the door") in association with the registration date and time T130-2-3 "2017/10/05 --08: 06: 58". One determination result T130-1 is stored. Further, in association with the registration date and time T130-2-4 "2017/10 / 05 --08: 07: 00", the first judgment result T130 indicating that the unit operation ID is "A3" (that is, "movement / walking"). -1 is stored. These unit movements are the registration date and time T130-2-3 ". It means that the subject opened the door and moved at "2017/10/05 --08: 06: 58". The subject took 56 seconds from the registration date and time T130-2-2 "2017/10 / 05 --08: 06: 02" to the registration date and time T130-2-3 "2017/10 / 05 --08: 06: 58". It may have taken some other action (such as looking for a key) in front of the same door. Alternatively, he may have moved through the hallway inside the house, and 56 seconds after starting the movement in the hallway, he opened the front door of the house and moved out of the front door.

After that, again, in association with the registration date and time T130-2-5 "2017/10/05 --08:10:00", the first judgment result T130 indicating that the unit operation ID is "A1" (that is, "open the door"). -1 is stored. At that time, it was associated with the above-mentioned registration date and time T130-2-1 "2017/10/05 --08:06:00" or registration date and time T130-2-4 "2017/10/05 --08:07:00". Assuming that the effective time is set to 1.5 minutes, the first determination result T130-1 is deleted from the first determination result table T130A by the effective time management process (S104A). As a result, the first determination result deleted from the first determination result table T130A is excluded from the determination target of the second determination process (S103). In the example of FIG. 6, it is associated with the registration date and time T130-2-1 "2017/10/05 --08:06:00" or the registration date and time T130-2-4 "2017/10/05 --08:07:00". In the series of first judgment results T130-1, the subject once opened the front door of the house and went outside, but noticed that he had forgotten the key and re-entered the house to pick up the key. It can be said that it is highly possible that it was due to the return. In other words, a series associated with the registration date and time T130-2-1 "2017/10/05 --08:06:00" or the registration date and time T130-2-4 "2017/10/05 --08:07:00". The first determination result T130-1 of the above is an example of the first determination result T130-1 which can be regarded as not constituting the behavior of a highly meaningful subject to be detected.

Next, in the example of FIG. 6, the unit operation ID indicates "A3" (that is, "walking / moving") in association with the registration date / time T130-2-6 "2017/10/05 --08: 10: 02". The first determination result T130-1 is stored. Further, in association with the registration date and time T130-2-7 "2017/10/05 --08: 10: 04", the first determination result T130-1 indicating that the unit operation ID is "A4" (that is, "insert") is stored. Has been done. Further, in association with the registration date and time T130-2-8 "2017/10 / 05 --08: 10: 05", the first judgment result T130 indicating that the unit operation ID is "A6" (that is, "twist the wrist"). -1 is stored. Then, in association with the registration date and time T130-2-9 "2017/10/05 --08: 10: 06", the first determination result T130-1 indicating that the unit operation ID is "A5" (that is, "pull out") is stored. Has been done. These unit operations are only a few from the registration date and time T130-2-5 "2017/10/05 --08:10:00" to the registration date and time T130-2-9 "2017/10/05 --08:10:06". Since it was detected in 6 seconds, if the valid time is set to 1.5 minutes, it is not deleted from the first determination result table T130A by the valid time management process (S104A). As a result, in the 6 seconds from the registration date and time T130-2-5 "2017/10/05 --08:10:00" to the registration date and time T130-2-9 "2017/10/05 --08:10:06" The judgment target of the second judgment process (S103) can be narrowed down to the first judgment result T130-1 after the registration date and time T130-2-5 "2017/10 / 05 --08: 10:00".

According to one aspect of the second embodiment disclosed above, in the second determination process (S103), the first determination result, which is the determination target of matching with the second pattern T120-3, is set as a time frame based on the effective time. Can be limited by. Such an action is useful for suppressing the waste of calculation cost when determining the advanced meaning of the behavior of the subject wearing the wearable device, and for more efficiently using the hardware resource.

According to another aspect of the second embodiment disclosed above, the second determination process (S103) is performed by limiting the first determination result, which is the determination result of the first determination process, to the time frame based on the effective time. ) Can suppress the occurrence of erroneous judgment. Such an action is useful for suppressing the waste of calculation cost when determining the advanced meaning of the behavior of the subject wearing the wearable device, and for more efficiently using the hardware resource.

<Example 3> In the behavior determination system 10 according to the third embodiment, the effective time in the effective time management means P140A is adjusted based on the measurement data from the sensor mounted on the subject.

FIG. 9 is a diagram showing an example of the configuration of the behavior determination system 10 according to the third embodiment. The behavior determination system 10 according to the third embodiment shown in FIG. 9 is different from the behavior determination system 10 according to the second embodiment shown in FIG. 5 in that the adjusting means P150B is added, and is the same in other respects. .. In the example of FIG. 9, the first adjusting means P160B is shown as an example of the processing related to the adjusting means P150B. In FIG. 9, the same reference numerals are given to the same configurations as those in FIG. 5, and the description of the same configurations will be omitted for the sake of brevity.

The adjusting means P150B is configured to adjust the effective time in the effective time management means P140A based on the measurement data from the sensor mounted on the subject.

The first adjusting means P160B is an example of a process (may be referred to as an adjusting process) related to the adjusting means P150B, and for example, the first operation is performed based on the measurement data from the motion sensor 130 mounted on the subject. When it is detected, the effective time in the effective time management means P140A is adjusted so as to be extended.

FIG. 10 is a diagram showing an example of the flow of the effective time management process (S104A) according to the third embodiment. The processing flow shown in FIG. 10 is different from the processing flow according to the second embodiment shown in FIG. 8 in that S104A-5B and S104A-6B are added, and is the same in other respects. In the following, for the sake of brevity, the description of S104A-5B and S104A-6B will be mainly described, and the description of other processes will be omitted. An example of the processing flow in the behavior determination system 10 according to the third embodiment is the same as the processing flow of the second embodiment shown in FIG. 7.

For example, the processor 110 executes the effective time management process (S104A) after executing the first determination process (S102), so that the process flow shown in FIG. 10 is executed. When the processor 110 repeatedly executes the process flow shown in FIG. 7 at predetermined time intervals, the effective time management process (S104A) is repeatedly executed at predetermined time intervals, and the process flow shown in FIG. 10 is repeatedly executed. You may.

The processor 110 determines whether or not the first determination result indicating the unit operation determined by the first determination process (S102) does not correspond to the predetermined first operation (S104A-5B). Here, the first motion is an motion in which the measurement data from the motion sensor 130 mounted on the subject indicates an exercise intensity equal to or higher than a predetermined threshold value, and is the first motion registered in the first pattern DB (T110). It is an operation that does not correspond to other unit operations other than. In other words, it can be said that it is not in a stationary state. For example, the action of taking out the key from the pocket of the trousers or the action of taking out the key from the bag may correspond to the first action. Such a first operation may be excluded from the determination target by the second determination process (S103) described later, but paying attention to the aspect that the operation is acceptable in the action determination system 10, for example, it is called an allowable operation. You may.

FIG. 11 is a diagram showing a content example of the first pattern DB (T110) according to the third embodiment. The first pattern DB (T110) shown in FIG. 11 stores the first definition information T110-4B whose unit operation ID (T110-1) is “A0”, and is the first according to the first embodiment shown in FIG. It differs from the content example of the pattern DB (T110) and is the same in other respects.

In the example of FIG. 11, in the first definition information T110-4B, the unit operation ID (T110-1) is "A0", the unit operation label T110-2 is "first operation", and the first pattern T110- 3 is the "feature amount Z". The feature amount Z may be any feature amount that can identify the above-mentioned first operation. For example, any data may be used as long as it can identify whether or not the measurement data from the motion sensor 130 is an operation indicating an exercise intensity equal to or higher than a predetermined threshold value.

In S104A-5B, if the unit operation ID shown in the first determination result does not match "A0" indicating the first operation, the processor 110 may determine that the first determination result is not the first operation ( YES with S104A-5B). On the other hand, in S104A-5B, when the unit operation ID shown in the first determination result matches "A0" indicating the first operation, the processor 110 determines that the first determination result is the first operation. It may be good (NO in S104A-5B).

When the processor 110 determines that the first determination result is not the first operation (YES in S104A-5B), the processor 110 may execute the processing after S104A-1 as in the second embodiment.

On the other hand, when the processor 110 determines that the first determination result is the first operation (NO in S104A-5B), the processor 110 extends the effective time (S104A-6B). For example, the processor 110 may extend the effective time in S104A-6B by adding a predetermined time length to the preset effective time. The predetermined time length to be added to the effective time may be set based on, for example, a time interval (execution interval, which may be referred to as a unit time) in which the processor 110 repeatedly executes the processing flow shown in FIG. 7. .. For example, when the unit time is 1 second, the processor 110 may use the value obtained by adding 1 second to the effective time as a new effective time (may be referred to as an updated effective time) in S104A-6B. good. As a result, when it is determined that the first judgment result is not the first operation (YES in S104A-5B) when the next valid time management process (S104A) is executed, the first judgment result is used using the updated valid time. Is searched (S104A-2), and the first determination result that exceeds the valid time is deleted from the first determination result table T130A (S104A-4).

As a modification of the above-mentioned S104A-6B, the processor 110 may extend the effective time by stopping the counting of the timer used for determining whether or not the effective time has elapsed. In this case, if the processor 110 determines that the first determination result is not the first operation (YES in S104A-5B), the processor 110 may execute the processing after S104A-1 and restart the timer counting.

According to one aspect of the third embodiment disclosed above, the effective time in the effective time management means P140A is adjusted based on the measurement data acquired from the sensor. Therefore, the effective time of the first determination result can be adjusted according to the situation grasped based on the measurement data from the sensor, and the occurrence of erroneous determination in the second determination process (S103) can be further suppressed. can. Such an action is useful for suppressing the waste of calculation cost when determining the advanced meaning of the behavior of the subject wearing the wearable device, and for more efficiently using the hardware resource.

According to another aspect of the third embodiment disclosed above, the effective time in the effective time management means P140A is extended during the period in which the first motion having a predetermined exercise intensity that is not in a stationary state is detected. .. Therefore, even if the first movement (which may be referred to as an extra movement or an unexpected movement) different from the series of unit movements described in the second pattern T120-3 appears in the movement of the subject. It is possible to prevent the valid time of the first determination result from expiring due to the onset period of the first action, and to prevent the subject's behavior from being missed.

According to still another aspect of Example 3 disclosed above, when a subject's behavior is defined by a combination of one or more unit movements, the minor movement is classified as the first movement. The amount of information of the second pattern T120-3 that defines the behavior of the body can be compressed. Such an action is useful for suppressing the waste of calculation cost when determining the advanced meaning of the behavior of the subject wearing the wearable device, and for more efficiently using the hardware resource.

<Example 4> In the behavior determination system 10 according to the fourth embodiment, another example of the adjusting means P150B for adjusting the effective time in the effective time management means P140A based on the measurement data from the sensor mounted on the subject is used. Provided.

FIG. 12 is a diagram showing an example of the configuration of the behavior determination system 10 according to the fourth embodiment. The behavior determination system 10 according to the fourth embodiment shown in FIG. 12 has the second adjusting means P170C as an example of the processing (which may be referred to as the adjusting processing) related to the adjusting means P150B. The behavior determination system 10 according to Example 2 is different from the behavior determination system 10 according to Example 3 shown in FIG. In the example shown in FIG. 12, the behavior determination system 10 according to the fourth embodiment includes the GPS (Global Positioning System) module 140C, but the GPS module 140C may be omitted. In FIG. 12, the same reference numerals as those shown in FIGS. 5 and 9 are designated by the same reference numerals, and detailed description thereof will be omitted for the sake of brevity.

The second adjusting means P170C is an example of the process relating to the adjusting means P150B, and is effective when, for example, the moving distance of the subject after the first determination result is registered in the first determination result table T130A is equal to or greater than the threshold value. The time management means P140A is adjusted so that the effective time is shortened.

The GPS module 140C is a GPS receiver capable of positioning the current position on the earth based on a signal from a GPS satellite, and is configured to be able to supply the current position as measurement data to the processor 110. In other words, in the present disclosure, the GPS module 140C is a sensor (position sensor) configured to output measurement data indicating the current position based on a signal from a GPS satellite (which may be referred to as an external signal). It may be referred to as).

13 and 14 are diagrams showing an example of the flow of the effective time management process (S104A) according to the fourth embodiment. The processing flow shown in FIGS. 13 and 14 is different from the processing flow according to the second embodiment shown in FIG. 8 in that S104A-7C to S104A-11C are added, and is the same in other respects. .. In the following, for the sake of brevity, the description of S104A-7C to S104A-11C will be mainly described, and the description of other processes will be omitted. An example of the processing flow in the behavior determination system 10 according to the fourth embodiment is the same as the processing flow of the second embodiment shown in FIG. 7.

For example, the processor 110 executes the effective time management process (S104A) after executing the first determination process (S102), so that the process flow shown in FIGS. 13 and 14 is executed. When the processor 110 repeatedly executes the processing flow shown in FIG. 7 at predetermined time intervals, the effective time management process (S104A) is repeatedly executed at predetermined time intervals, and the processing flow shown in FIGS. 13 and 14 is repeated. It may be executed repeatedly.

The processor 110 determines whether the first determination result indicating the unit operation determined by the first determination process (S102) in the current execution is the same as the previous first determination result (S104A-7C). In S104A-7C, the processor 110 compares the latest first judgment result information registered in the first judgment result table T130A with the first judgment result of this time, so that the first judgment result of this time is the previous time. It may be determined whether or not it is the same as the first determination result of. For example, when the processor 110 matches the first determination result T130-1 shown in the latest first determination result information acquired from the first determination result table T130A with the current determination result, the processor 110 is the current first determination result. May be determined to be the same as the previous first determination result (NO in S104A-7C). On the other hand, when the first determination result T130-1 shown in the latest first determination result information acquired from the first determination result table T130A and the present determination result do not match, the processor 110 is the present first determination result. May be determined not to be the same as the previous first determination result (YES in S104A-7C).

When it is determined in S104A-7C that the current first determination result is not the same as the previous first determination result (YES in S104A-7C), the processor 110 initializes the moving distance of the subject (S104A-8C). ). In S104A-8C, the processor 110 may set, for example, the travel distance of the subject to a zero value.

Similar to the second embodiment, the processor 110 registers the current first determination result in the first determination result table T130A together with the registration date and time (S104A-1). In the example of FIG. 13, it is shown that S104A-1 is executed after the execution of S104A-8C, but this embodiment is not limited to this execution order. For example, the execution order of S104A-8C and S104A-1 may be reversed.

On the other hand, when it is determined in S104A-7C that the first determination result of this time is the same as the first determination result of the previous time (NO in S104A-7C), the processor 110 initializes the moving distance (S104A-8C). ) May be skipped without executing. In the example of FIG. 13, when NO in S104A-7C, the process of registering the first determination result in the first determination result table (S104A-1) is not executed and is skipped.

Next, the processor 110 calculates the travel distance based on the measurement data from the sensor (S104A-9C). In S104A-9C, the processor 110 acquires, for example, the position information shown in the measurement data from the GPS module 140C (which may be referred to as a position sensor), and is based on the current position information and the previous position information. The travel distance may be calculated. In this case, each time the processor 110 executes S104A-9C, the processor 110 may store the position information (which may be referred to as the current position information) shown in the measurement data acquired from the GPS module 140C in the memory 120. good. As a result, the position information stored in the memory 120 can be used as the previous position information at the next execution of S104A-9C. The travel distance calculated in S104A-9C may be accumulated (may be referred to as a cumulative sum) until it is initialized in S104A-8C.

In S104A-9C, the processor 110 may calculate the movement distance of the subject based on the measurement data from the motion sensor 130, for example. The measurement data acquired from the motion sensor 130 may indicate, for example, a value obtained by measuring a change in body movement caused by movement such as walking of the subject. Based on the measurement data from the motion sensor 130, the processor 110 detects that the subject is in the walking / moving state by a known method, and indicates the moving distance based on the time during which the walking / moving state continues. The index value may be calculated. Alternatively, the processor 110 may measure the number of steps of the subject by a known method based on the measurement data from the motion sensor 130, and calculate an index value indicating the distance traveled based on the measured number of steps.

The processor 110 determines whether the travel distance is equal to or greater than the threshold value (S104A-10C). For example, the processor 110 compares the accumulated travel distance with a predetermined threshold value each time S104A-9C is executed, and if the travel distance is a value equal to or greater than the threshold value, determines that the travel distance is equal to or greater than the threshold value. It may be good (YES in S104A-10C). On the other hand, the processor 110 compares, for example, the accumulated movement distance each time S104A-9C is executed with a predetermined threshold value, and if the movement distance is a value less than the threshold value, determines that the movement distance is not equal to or more than the threshold value. It may be (NO in S104A-10C).

Here, the predetermined threshold value to be compared with the movement distance may be set to any value among the values exceeding the movement distance that can be expressed in a series of unit movements constituting the behavior that the subject can take. In other words, the threshold value may be set to a value at which it can be determined that the movement distance that can be expressed in the combination of a series of unit movements constituting the behavior that the subject can take is clearly exceeded. However, the larger the value is set, the more the first judgment result included in the judgment target of the second judgment process can be, so that the effect of suppressing the waste of the calculation cost required for the behavior judgment of the subject can be reduced. Please note. Therefore, from the viewpoint of suppressing the waste of calculation cost, among the values that can clearly determine that the travel distance that can be expressed in the combination of a series of unit movements constituting the behavior that the subject can take is exceeded, as much as possible. It is recommended to set a small value as the threshold.

When it is determined in S104A-10C that the moving distance is equal to or greater than the threshold value (YES in S104A-10C), the processor 110 shortens the effective time (S104A-11C). For example, the processor 110 may shorten the effective time by setting the effective time to a value of 0. Alternatively, the processor 110 may shorten the valid time by subtracting any value other than the zero value (which may be referred to as a positive number) from the valid time.

On the other hand, in S104A-10C, when it is determined that the moving distance is less than the threshold value (NO in S104A-10C), the processor 110 may skip without executing the process of shortening the effective time (S104A-11C). good.

After that, the processor 110 executes the processing after S104A-2 as in the second embodiment. As a result, the first judgment result is searched using the effective time shortened according to the judgment result of S104A-10C (S104A-2), and the first judgment result exceeding the valid time is obtained from the first judgment result table T130A. It is deleted (S104A-4).

According to one aspect of the fourth embodiment disclosed above, the effective time in the effective time management means P140A is adjusted based on the measurement data acquired from the sensor. Therefore, the effective time of the first determination result can be adjusted according to the situation grasped based on the measurement data from the sensor, and the occurrence of erroneous determination in the second determination process (S103) can be further suppressed. can. Such an action is useful for suppressing the waste of calculation cost when determining the advanced meaning of the behavior of the subject wearing the wearable device, and for more efficiently using the hardware resource.

According to another aspect of Example 4 disclosed above, when it is detected that the moving distance of the subject is equal to or greater than the threshold value, the effective time in the effective time management means P140A is shortened. As a result, the valid time of the first determination result can be easily expired, and the first determination result stored in the memory 120 can be excluded from the determination target of the second determination process (S103). In other words, when it is detected that the movement distance that can be expressed in the combination of a series of unit movements constituting the behavior that the subject can take is clearly exceeded, the already accumulated first judgment result is the second judgment. It is excluded from the determination target of the process (S103). Such an action is useful for suppressing the waste of calculation cost when determining the advanced meaning of the behavior of the subject wearing the wearable device, and for more efficiently using the hardware resource.

<Embodiment 5> In the action determination system 10 according to the fifth embodiment, preset processing (action) is performed according to the determination result (may be referred to as a second determination result, an event, etc.) of the second determination means P130. May be referred to as) is executed.

FIG. 15 is a diagram showing an example of the configuration of the behavior determination system 10 according to the fifth embodiment. The behavior determination system 10 according to the fifth embodiment shown in FIG. 15 has at least the event processing means P180D and the action setting table T140D. It is different from the configuration according to 2, the configuration according to the third embodiment illustrated in FIG. 9, and the configuration according to the fourth embodiment illustrated in FIG. In the example shown in FIG. 15, the behavior determination system 10 according to the fifth embodiment includes the communication circuit 150D, but the communication circuit 150D may be omitted. Further, in the example shown in FIG. 15, the behavior determination system 10 according to the fifth embodiment has the first adjusting means P160B and the second adjusting means as an example of the processing (may be referred to as the adjusting processing) of the adjusting means P150B. It has P170C, but the present embodiment is not limited to this. Further, in the example shown in FIG. 15, the event log T150D is shown in the action determination system 10 according to the fifth embodiment, but the event log T150D is also provided in the action determination system 10 according to another embodiment. It may be omitted or it may be omitted. In FIG. 15, the configuration according to the first embodiment illustrated in FIG. 1, the configuration according to the second embodiment exemplified in FIG. 5, the configuration according to the third embodiment exemplified in FIG. 9, and the fourth example illustrated in FIG. The same reference numerals are given to the same configurations as those of the above configurations, and detailed description thereof will be omitted for the sake of brevity.

In FIG. 15, the event processing means P180D is configured to acquire and execute an action corresponding to an action that the subject can take from the action setting table T140D.

The action setting table T140D is information about an action (which may be referred to as action setting information) which is a process to be executed depending on whether or not an action that the subject can take, which is determined by the second determination means P130, is detected. Is configured to be set.

The event log T150D is configured so that the information regarding the behavior of the subject determined by the second determination means P130 is recorded together with the information regarding the date and time when the behavior of the subject is detected.

The communication circuit 150D is, for example, a wireless communication standard (for example, LTE (Long Term Evolution) standard, 5G) established by a standardization body (for example, 3GPP (3rd Generation Partnership Project), IEEE (The Institute of Electrical and Electronics Engineers), etc.). It may be a wireless communication device compliant with standards (standards, IEEE802.11 standards, etc.).

FIG. 16 is a diagram showing a content example of the action setting table T140D according to the fifth embodiment. In FIG. 16, the action setting table T140D includes an action ID (T140-1) that uniquely identifies the action setting information, a condition 1 (T140-2) and a condition 2 (T140-3) that are one or more conditions. The action setting information including the action T140-4, which is a process executed when the condition is fulfilled, is stored.

In the example of FIG. 16, the action setting information of the action ID (T140-1) "C1" is fulfilled when it is detected that "they have left the radius [50 m] from the home" as the condition 1 (T140-2). When the condition is set and the condition 2 (T140-3) is fulfilled when "action ID [B1] is not detected", and both condition 1 and condition 2 are fulfilled. It is set to "output a warning" as the action T140-4 to be executed. Here, the action ID [B1] corresponds to the action of "locking the front door".

Further, in the example of FIG. 16, the action setting information of the action ID (T140-1) “C2” is set as the condition 1 (T140-2) to be fulfilled when “the action ID [B3] is detected”. As condition 2 (T140-3), a condition to be fulfilled when "the current time is within the range of [17: 30-23: 59]" is set, and both condition 1 and condition 2 are set. As the action T140-4 to be executed when it is fulfilled, it is set to "send [mail 1] to [destination 1]". Here, the action ID [B3] corresponds to the action of "starting driving the car".

FIG. 17 is a diagram showing an example of a processing flow in the behavior determination system 10 according to the fifth embodiment. The process flow according to the fifth embodiment shown in FIG. 17 is the process flow according to the first embodiment shown in FIG. 4 and FIG. 7 in that at least the step (S106D) for executing the event process is added. It is different from the flow of the process according to the second embodiment shown in the above. Further, in the processing flow according to the fifth embodiment shown in FIG. 17, a step (S105D) for outputting the determination result (which may be referred to as the second determination result) of the second determination process S103 to the event log is specified. In that respect, the flow of processing according to Example 1 shown in FIG. 4 and the flow of processing according to Example 2 shown in FIG. 7 are different. In the following, for the sake of brevity of explanation, the same reference numerals are given to the steps similar to the processing flow according to the first embodiment shown in FIG. 4 and the processing flow according to the second embodiment shown in FIG. A detailed description will be omitted.

For example, the processor 110 executes the effective time management process (S104A) after executing the first determination process (S102), so that the first determination result indicating the unit operation determined by the first determination process (S102) is shown. Is registered in the first determination result table T130A. In S104A, the processor 110 manages the valid time of each of the one or more first judgment results registered in the first judgment result table T130A, and the first judgment result after the valid time has passed is obtained from the first judgment result table T130A. delete. The effective time management process (S104A) may be omitted in the process flow according to the fifth embodiment.

In the second determination process (S103), the processor 110 determines an action that the subject can take based on one or more valid first determination results registered in the first determination result table T130A. The determination result by the second determination process (S103) may be referred to as a second determination result. In other words, the second determination result indicates an action that a subject determined based on one or more valid first determination results can take.

In S103, the processor 110 may determine that the first determination result whose elapsed time from being registered in the first determination result table T130A does not exceed the effective time is a valid first determination result. Alternatively, in S103, the processor 110 may determine that the first determination result that has not been deleted from the first determination result table T130A by the effective time management process (S104A) is a valid first determination result.

The processor 110 outputs (stores) the information regarding the second determination result to the event log T150D together with the date and time regarding the second determination result (S105D).

FIG. 18 is a diagram showing a content example of the event log T150D according to the fifth embodiment. In the example of FIG. 18, the event log T150D has an event ID (T150-1), a second determination result ID (T150-2), a second determination result label T150-3, and an event date / time T150-4. Store event information.

The event ID (T150-1) is information used to uniquely identify event information, and may be described using numbers, letters, symbols, or a combination thereof.

The second determination result ID (T150-2) is information that uniquely identifies the behavior of the subject indicated by the second determination result, and corresponds to the behavior ID (T120-1) in the second pattern DB (T120). ..

The second determination result label T150-3 is information describing the behavior of the subject indicated by the second determination result, and corresponds to the behavior label T120-2 in the second pattern DB (T120).

The event date and time T150-4 is information indicating the date and time when the event information is registered in the event log.

In the example of FIG. 18, in the event information of the event ID (T150-1) "D1", "B1" is set in the second determination result ID (T150-2), and the second determination result label T150-3 is set. "Lock the front door" is set, and "2017/10/05 --08:10:06" is set for the event date and time T150-4. In other words, the event information of the event ID (T150-1) "D1" is based on one or more unit actions, and the subject's action of "locking the front door" is "2017/10 / 05-". 08:10:06 ”indicates that the judgment was made.

Further, in the example of FIG. 18, in the event information of the event ID (T150-1) "D2", "B3" is set in the second determination result ID (T150-2), and the second determination result label T150- "Start driving the car" is set in 3, and "2017/10 / 05 --08: 13: 10" is set in the event date and time T150-4. In other words, the event information of the event ID (T150-1) "D2" is based on one or more unit movements, and the subject's behavior of "starting driving" is "2017/10 / 05 --08". : 13:10 "indicates that the judgment was made.

Further, in the example of FIG. 18, in the event information of the event ID (T150-1) "D3", "B3" is set in the second determination result ID (T150-2), and the second determination result label T150- "Start driving the car" is set in 3, and "2017/10 / 05 --21: 43: 40" is set in the event date and time T150-4. In other words, the event information of the event ID (T150-1) "D3" is based on one or more unit movements, and the subject's behavior of "starting driving" is "2017/10 / 05-21. : 43: 40 "indicates that the judgment was made.

Returning to the description of the processing flow according to the fifth embodiment shown in FIG. The processor 110 executes event processing (S106D). In S106D, the processor 110 acquires and executes the action T140-4 corresponding to the action that the subject can take from the action setting table T140D based on the second determination result.

In S106D, the processor 110 acquires, for example, the action setting information from the action setting table T140D, and determines whether or not the condition shown in the action setting information is satisfied based on the second determination result. If necessary, the processor 110 may determine whether or not the condition shown in the action setting information is satisfied based on the measurement data from the motion sensor 130 or the GPS module 140C, for example.

In the example of FIG. 16, the processor 110 measures from the GPS module 140C whether or not the condition 1 (T140-2) whose action ID (T140-1) is shown in the action setting information of “C1” is satisfied. May be determined based on. For example, the processor 110 calculates the distance between the position preset as the home position and the current position indicated by the measurement data from the GPS module 140C, and if the distance exceeds 50 m, the action ID (T140). It may be determined that -1) fulfills the condition 1 (T140-2) shown in the action setting information of "C1".

If it is determined that the condition 1 (T140-2) indicated in the action setting information of "C1" for the action ID (T140-1) is satisfied, the processor 110 determines whether or not the condition 2 (T140-3) is satisfied. May be determined. For example, the processor 110 refers to the event log T150D, and among the event information having the event date and time T150-4 within a predetermined time (for example, within 30 minutes) from the current time, the second determination result ID (T150-2) is It may be determined whether or not the event information indicating the action ID "B1" is stored in the event log T150D. If such event information is not stored in the event log T150D, the processor 110 may determine that the condition 2 (T140-3) that "action ID [B1] has not been detected" has been fulfilled.

In the example of FIG. 16, the processor 110 determines that the condition 1 (T140-2) and the condition 2 (T140-3) whose action ID (T140-1) is shown in the action setting information of “C1” are fulfilled. In this case, the process of "outputting a warning" set in the action T140-4 is executed. In the process of outputting a warning, the processor 110 may, for example, output a warning sound from an acoustic output circuit (which may be referred to as a speaker) (which may be referred to as a speaker) (not shown in FIG. 15), or a display circuit (display) which is not shown in FIG. A warning message may be output. As a result, when leaving the predetermined area, a warning can be output depending on whether or not the second determination result indicating the predetermined behavior of the subject has been detected. For example, by outputting a warning, it may be possible to give an opportunity for the subject to re-perform a predetermined action, for example. In this embodiment, the person who receives the warning is not limited to the subject. For example, it may be a person other than the subject (may be referred to as a subject's manager, caregiver, caregiver, guardian, owner, breeder, etc.).

An example of another action setting information shown in FIG. 16 will be described. In the example of FIG. 16, the processor 110 determines whether or not the condition 1 (T140-2) whose action ID (T140-1) is shown in the action setting information of “C2” is satisfied based on the second determination result. , May be determined. For example, in the processor 110, when the action of the subject indicated by the second determination result corresponds to the action ID [B3], the condition 1 in which the action ID (T140-1) is shown in the action setting information of "C2". It may be determined that (T140-2) is fulfilled.

If it is determined that the condition 1 (T140-2) indicated in the action setting information of "C2" for the action ID (T140-1) is satisfied, the processor 110 determines whether or not the condition 2 (T140-3) is satisfied. May be determined. For example, the processor 110 may determine that condition 2 (T140-3) is fulfilled if the current time is within the range [17: 30-23: 59].

In the example of FIG. 16, the processor 110 determines that the condition 1 (T140-2) and the condition 2 (T140-3) whose action ID (T140-1) is shown in the action setting information of “C2” are fulfilled. In this case, the process of "sending [mail 1] to [destination 1]" set in the action T140-4 is executed. In executing the action T140-4, the processor 110 may send a preset text e-mail [mail 1] to a preset [destination 1] using, for example, the communication circuit 150D. good. As a result, in the example of FIG. 18, the conditions in the action setting information in which the event ID (T150-1) is "D2" and "D3" and the action ID (T140-1) is "C2". Condition 2 is satisfied because 1 (T140-2) is satisfied, but the event date and time T150-4 in the event information whose event ID (T150-1) is "D2" is "2017/10 / 05 --08: 13: 10". (T140-3) is not satisfied. On the other hand, since the event date and time T150-4 in the event information whose event ID (T150-1) is "D3" is "2017/10 / 05 --21: 43: 40", the condition 2 (T140-3) is satisfied. As a result, according to the second determination result that triggered the event information whose event ID (T150-1) is "D3" to be registered in the event log T150D, the processor 110 sends "[mail 1] to [destination 1]. ] Is executed. For example, by setting the e-mail address of someone in the family in [Destination 1] and setting the message "I'm going home by car" in [Email 1], the second judgment result will be obtained. Based on this, it is possible to automatically convey a predetermined message to the family.

According to one aspect of the fifth embodiment disclosed above, a preset process (which may be referred to as an action) is executed according to the second determination result by the second determination means P130. Such an action more efficiently determines the advanced meaning of the behavior of the subject wearing the wearable device, and whether or not to execute a preset process according to the advanced meaning of the behavior of the subject. It is useful for realizing a system that can control the behavior.

<Modification 1> According to the behavior determination system 10 according to the first to fifth embodiments disclosed above, a specific example of the arrangement of each component such as the processor 110, the memory 120, and the motion sensor 130 is particularly mentioned. Not done. For example, the component of the behavior determination system 10 according to the first embodiment shown in FIG. 1, the component of the behavior determination system 10 according to the second embodiment shown in FIG. 5, and the behavior determination system 10 according to the third embodiment shown in FIG. Each of the components, the component of the behavior determination system 10 according to the fourth embodiment shown in FIG. 12, and the component of the behavior determination system 10 according to the fifth embodiment shown in FIG. 15 is one device (for example, to the subject). It may be arranged in a wearable device to be attached), or it may be distributed in a plurality of devices.

For example, among the components of the behavior determination system 10 according to the fifth embodiment shown in FIG. 15, the motion sensor 130 is arranged on a wearable device to be attached to the subject, and the other components are information processing of a smartphone, a server, or the like. It may be placed in the device. In this case, the information processing device is directly or indirectly connected to the wearable device by wireless communication technology such as Bluetooth (registered trademark), wireless LAN (Local Area Network), or LTE, and is measured from the motion sensor 130. Data shall be able to be received from the wearable device.

Alternatively, among the components of the behavior determination system 10 according to the fifth embodiment shown in FIG. 15, for example, the processor 110 provided with the data acquisition means P110 or the second adjustment means P170C and the first pattern DB (T110) to the first. The memory 120 provided with the determination result table T130A, the motion sensor 130, and the GPS module 140C may be arranged in a wearable device to be attached to the subject. Further, for example, the processor 110 provided with the event processing means P180D, the memory 120 provided with the action setting table T140D and the event log T150D, the GPS module 140C, and the communication circuit 150D are combined with an information processing device (smartphone, server, etc.). ) May be placed. In this case, the information processing device is directly or indirectly connected to the wearable device by the wireless communication technology, and can receive information about the second determination result by the second determination means P130 from the wearable device.

<Modification 2> According to the behavior determination system 10 according to the first to fifth embodiments disclosed above, the first pattern DB (T110) and the second pattern DB (T120) are stored in the memory 120. However, the present invention is not limited to such embodiments. For example, it may be implemented as an algorithm or data in a program as various means executed by the processor 110. It should be noted that such an embodiment also belongs to the concepts of the first pattern DB (T110) and the second pattern DB (T120) in the present disclosure.

Further, the first determination result table T130A has been described as if the information elements exemplified in FIG. 6 are stored in the memory 120 as a block, but the present invention is not limited to such an embodiment. Please note. For example, each information element may have some relation and may be distributed and stored in the memory 120, or may be only temporarily stored in the processor 110 or the memory 120. Furthermore, it may be implemented as an algorithm or data in a program as various means executed by the processor 110. It should be noted that such an embodiment also belongs to the concept of the first determination result table T130A in the present disclosure. The same applies to various other tables.

The above detailed description will clarify the features and advantages of the embodiments. This is intended to extend to the features and advantages of the embodiments as described above, to the extent that the claims do not deviate from their spirit and scope of rights. Also, anyone with normal knowledge in the art should be able to easily come up with any improvements or changes. Therefore, there is no intention to limit the scope of the embodiments having the invention to the above-mentioned ones, and it is possible to rely on appropriate improvements and equivalents included in the scope disclosed in the embodiments. For example, each step disclosed herein does not necessarily have to be processed in chronological order according to the order described as an example of the processing flow, and is within the scope of the gist of the present invention described in the claims. In, the order of the steps may be changed, or a plurality of steps may be executed in parallel. The circumstances that may occur in the behavior determination system 10 clarified in the above detailed explanation can be found when the behavior determination system 10 is examined from one aspect, and when examined from another aspect, it can be found. Note that other circumstances can be found. In other words, the features and advantages of the present invention are not limited to the applications that fulfill the circumstances specified in the above detailed description.

Finally, the configuration of the embodiments of the present disclosure is an example for embodying the technical idea of the present invention, and is not intended to limit the present invention to the configuration of the present embodiment. , And other embodiments within the scope of the claim. For example, it should be noted that the terms in this disclosure may change their names due to changes in future technological trends. It should also be noted that one or more alternatives listed for the terms in this disclosure may be synonymous with each other.

10: Behavior determination system 110: Processor 120: Memory 130: Motion sensor 140C: GPS module 150D: Communication circuit P110: Data acquisition means P120: First determination means P130: Second determination means P140A: Effective time management means P150B: Adjustment means P160B: First adjusting means P170C: Second adjusting means P180D: Event processing means T110: First pattern DB
T110-1: Unit operation ID
T110-2: Unit operation label T110-3: First pattern T120: Second pattern DB
T120-1: Action ID
T120-2: Action label T120-3: Second pattern T130A: First judgment result table T130-1: First judgment result T130-2: Registration date and time T140D: Action setting table T140-1: Action ID
T140-2: Condition 1
T140-3: Condition 2
T140-4: Action T150D: Event log T150-1: Event ID
T150-2: Second judgment result ID
T150-3: Second judgment result label T150-4: Event date and time

Claims (6)

It is an action judgment system,
A data acquisition means for acquiring measurement data from a sensor mounted on a subject, and
A first determination means for determining the unit movement from the measurement data based on a unit feature amount for identifying a unit movement which is an individual movement constituting the behavior that the subject can take.
The first determination result indicating the unit operation determined by the first determination means is registered in the first determination result table, and the valid time of the first determination result registered in the first determination result table is managed. An effective time management means for deleting the first determination result after the effective time has passed from the first determination result table.
A second determination means for determining possible actions by the subject based on a combination of one or more of the unit movements indicated by one or more valid first determination results registered in the first determination result table. When,
The adjusting means for adjusting the effective time in the effective time management means, and the adjusting means.
Equipped with
The adjusting means extends the effective time in the effective time management means when the first operation excluded from the combination of the unit operations is detected in the determination by the second determination means based on the measurement data. The effective time management means in the effective time management means when the moving distance of the subject after the first determination result is registered in the first determination result table is equal to or greater than the threshold value. second adjusting means and the at least behavior determining system that either have a one to adjust so that the time is shortened. The behavior determination system according to claim 1.
An action setting table in which information about an action, which is a process to be executed depending on whether or not an action that the subject can take, which is determined by the second determination means, is detected, and an action setting table are set.
An event processing means and an event processing means for acquiring and executing the action corresponding to the action that the subject can take from the action setting table.
Behavior judgment system equipped with. A behavior determination method,
Performs a data acquisition process to acquire measurement data from a sensor mounted on the subject,
A first determination process for determining the unit movement from the measurement data is executed based on the unit feature amount for identifying the unit movement which is an individual movement constituting the behavior that the subject can take.
The first judgment result indicating the unit operation judged by the first judgment process is registered in the first judgment result table, and the valid time of the first judgment result registered in the first judgment result table is managed. An effective time management process for deleting the first determination result after the valid time has passed from the first determination result table is executed.
A second determination process for determining possible actions by the subject based on a combination of one or more of the unit movements indicated by one or more valid first determination results registered in the first determination result table. And execute
Executing the adjustment process for adjusting the effective time in the effective time management process.
Look at including it,
In the adjustment process, the effective time in the effective time management process is extended when the first operation excluded from the combination of the unit operations is detected in the determination by the second determination process based on the measurement data. When the movement distance of the subject after the first determination result is registered in the first determination result table is equal to or greater than the threshold value, the first adjustment process is adjusted so as to be effective in the effective time management process. A behavior determination method having at least one of a second adjustment process for adjusting so that the time is shortened. The behavior determination method according to claim 3.
Actions that the subject can take from the action setting table in which information about the action that should be executed depending on whether or not the action that the subject can take, which is determined by the second determination process, is set. Acquire and execute the above action corresponding to, execute event processing,
Behavior judgment method including that. It ’s an action judgment program.
An acquisition means and an acquisition means for acquiring measurement data from a sensor mounted on a subject.
A first determination means for determining the unit movement from the measurement data based on a unit feature amount for identifying a unit movement which is an individual movement constituting the behavior that the subject can take.
The first determination result indicating the unit operation determined by the first determination means is registered in the first determination result table, and the valid time of the first determination result registered in the first determination result table is managed. An effective time management means for deleting the first determination result after the effective time has passed from the first determination result table.
A second determination means for determining possible actions by the subject based on a combination of one or more of the unit movements indicated by one or more valid first determination results registered in the first determination result table. And ,
The adjusting means for adjusting the effective time in the effective time management means, and the adjusting means.
To operate the computer as,
The adjusting means extends the effective time in the effective time management means when the first operation excluded from the combination of the unit operations is detected in the determination by the second determination means based on the measurement data. The effective time management means in the effective time management means when the moving distance of the subject after the first determination result is registered in the first determination result table is equal to or greater than the threshold value. second adjusting means and, behavior determining program that have a at least one of adjusting so that the time is shortened. The action determination program according to claim 5.
Actions that the subject can take from the action setting table in which information about the action, which is a process to be executed depending on whether or not the action that the subject can take, which is determined by the second determination means, is set. Event processing means, which acquires and executes the action corresponding to
An action judgment program that operates a computer as.

Images