JP2014190724A - Space state determination device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a space state determination device for finally determining the state of a space on the basis of whether the time sequence of the space state or the environment of the space is contradictory to the typical state transition of the space.SOLUTION: A space state estimation part 12 has an SVM 12a as a learning model, and estimates the state of a space on the basis of a received wave. An environment sensor part 13 includes an environment sensor such as a temperature sensor 13a, a humidity sensor 13b and a CO2 sensor 13c. A behavior information collection part 14 acquires behavior information preliminarily registered on a cloud by a user, which is a managed schedule and the on/off state of a home electric appliance such as a television or a lighting apparatus. The estimation result of the space state, the environment sensor output and the behavior information are provided via a network NW to a state determination part 22. The state determination part 22 finally determines the state of the space on a time (state determination time) when the state determination is requested on the basis of the estimation result of the space state, the environment sensor output and the behavior information.

Description

  The present invention relates to a spatial state determination apparatus that determines a state of a space using a radio wave sensor and an environment sensor, and in particular, determines the true state of the space based on a time series of spatial states observed discretely. The present invention relates to a spatial state determination device.

  There is known an event detection device that detects an event (operation state) such as an intrusion, movement, or opening / closing of a door using an infrared beam sensor or a microwave. Japanese Patent Application Laid-Open No. 2004-133867 discloses a technique for detecting the direction of arrival of radio waves from radio signals received in a plurality of channels of a receiver having an array antenna and detecting behavior in a room.

  In Non-Patent Document 1, as information on an incoming wave having the highest reception sensitivity, a first eigenvector obtained from a correlation matrix of received signals of a plurality of channels is applied to an SVM (support vector machine), so that Techniques for detecting behavior are disclosed.

  In Patent Document 2, a radio wave sensor device that receives radio waves transmitted from a transmitter with a plurality of channels of an antenna array and determines a state of a space in which the radio waves propagate based on the received waves of each antenna array is discretely disclosed. A technique for determining the state of the space based on whether the time series of the state of the observed space is consistent with the typical state transition of the space is disclosed.

JP 2008-216152 A Japanese Patent Application No. 2012-33235

"State identification using support vector machine based on signal subspace for wireless monitoring system" IEICE technical report. RCS, Wireless communication systems 110 (127), 143-148, 2010-07-08

  SVM can discretely estimate the state (class) of the space, but even if the state of the space changes due to the movement of an object, for example, it can detect the state of the space taking into account the continuous change in that time series Can not. Therefore, in the radio wave sensor device adopting SVM, it is particularly difficult to accurately determine the state in which the object moves.

  For example, when a radio wave sensor device is applied to watch over a room to determine "absence", "in-room (operating)" and "in-room (stationary)", the person is still in the room but is still There is a large difference in characteristics between the received wave detected by the radio wave sensor between the state and the absence state where the user is not in the room.

  However, for example, in order to transition from the “in-room (stationary)” state to the “absence” state, a person must move through the room to the entrance during that time, so if the monitoring cycle is about 1 second It is impossible for the state to transit directly from “in-room (stationary)” to “absence”, and the “in-room (motion)” state should be observed during that time. Therefore, even if a direct transition from “in-room (stationary)” to “absence” is observed, it can be recognized that there is a high possibility that “absence” is an error in “in-room (stationary)”.

  Furthermore, if the indoor CO2 concentration is high, there is a high possibility that the room is in the room, and if the room temperature is too high or too low for the general comfortable temperature, it is likely that the room is absent. Considering the environmental state, it becomes possible to determine the spatial state with higher accuracy.

  The object of the present invention is to solve the above-mentioned problems of the prior art, based on whether the time series of spatial states observed in a discrete manner and the environment of the space are consistent with typical state transitions of the space. An object of the present invention is to provide a space state determination apparatus that finally determines the state of the space.

  In order to achieve the above object, the present invention receives a radio wave transmitted from a transmitter by an antenna array of a plurality of channels, and determines a state of a space in which the radio wave propagates based on a received wave of each antenna array. The state determination device is characterized in that the following measures are taken.

  (1) Feature quantity calculation means for periodically calculating the feature quantity of received waves detected by each antenna array, and spatial state estimation for estimating the state of the space by applying the feature quantity of each period to a predetermined learning model Means, an environmental sensor for detecting the environment of the space, and a space state determination means for determining the state of the space at a predetermined state determination time based on the estimation result of the space state at the time and the output of the environment sensor. did.

  (2) It further comprises action information acquisition means for acquiring action information of a person acting in space, and the space state determination means indicates the state of the space at a predetermined state determination time, the estimation result of the space state at the time, the environment The determination was made based on the sensor output and behavior information.

  According to the present invention, the following effects are achieved.

  (1) The state of the space is periodically estimated by applying the feature value obtained from the received waves detected by the antenna array of multiple channels to the learning model, and the environment of the space is also detected, and their time series From this, the state of the space is determined. Therefore, if the time series of observed states does not match the typical state transition of the space or indicates a state transition that is not logically possible, the estimation by the learning model is incorrect. Therefore, it is possible to prevent state determination based on an erroneous estimation result and improve the accuracy of state determination.

  (2) In addition to spatial state estimation results and spatial environment, it is possible to exclude state transitions that are logically impossible as human behavior from the determination results by determining the spatial state in consideration of human behavior information. Therefore, the determination accuracy of the space state can be further improved.

It is the figure which showed the example of application of the space state determination apparatus which concerns on this invention. It is the functional block diagram which showed the structure of the space state determination apparatus to which this invention is applied. It is a functional block diagram of one embodiment of a state judgment part. It is the figure which showed an example of history information DB. It is the figure which showed an example of the conditional expression table. It is the figure which showed an example of the determination rule table. It is a functional block diagram of other embodiments of a state judgment part. It is the figure which showed an example of the Bayesian network. It is the figure which showed the case of the CO2 sensor as an example the posterior probability table | surface which calculates | requires environment determination result S ^ j, t from environmental sensor output Sj, t. It is the figure which showed the posterior probability table | surface which calculates | requires action determination result I ^ k, t from action information Ik, t using the schedule (meeting) as an example. Spatial state judgment result X ^ based on spatial state estimation result Xt, environmental judgment result S ^ j, t, action judgment result I ^ k, t and past spatial state judgment (X ^ ti to X ^ t-1) It is the figure which showed an example of the posterior probability table | surface which calculates | requires t.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the present invention, if an array antenna having a plurality of arrays is adopted as the antenna structure of the receiver, characteristics such as the received radio wave intensity and the received radio wave time detected at the same time in each array are in the state of radio wave interference. Focusing on the particularity of the received radio wave characteristics, the relationship between the received radio wave characteristics and the state of the space that changes according to the position and orientation of the radio wave interferer is learned and modeled in advance. Is applied to the learning model to estimate the state of the space.

  Furthermore, in this embodiment, in addition to such a spatial state estimation, an environment that changes depending on whether or not a person is present in the space is detected by various environmental sensors, and further, a person's action schedule, action results, etc. The result of the estimation of the spatial state, the environmental detection by the environmental sensor and the monitoring of the behavior are periodically repeated, and the time series is applied to a knowledge base such as a predetermined rule or established model to finalize the spatial state. judge.

  FIG. 1 is a diagram showing an application example of a space state determination device using an array antenna. Here, a case where a person to be monitored is regarded as a radio wave interference object and the presence or behavior of a person in a room is determined. An example will be described. In this embodiment, in addition to the antenna array, various environmental sensors such as a temperature sensor, a humidity sensor, and / or a CO2 sensor are disposed at appropriate locations inside and outside the room (not shown), and further, home appliances such as a television and a lighting fixture. The on / off state and the user's own schedule registered on the cloud are acquired as behavior information.

  When the transmitter Tx (lower right in the figure) and the array antenna 10 (upper left in the figure) are arranged at appropriate locations in the room, the radio waves transmitted from the transmitter Tx can be directly or fixedly attached to furniture or walls or human beings. It is reflected by such a moving body and reaches the array antenna 10 to be received. At this time, if a person is not present in the room, the propagation path of the radio wave does not change, so that the characteristics of the radio wave received by each array of the array antenna 10 do not change greatly. Further, even when a person is present, if the radio wave propagation path does not change if the user is in a stationary state such as sleeping, the radio wave received by each array of the array antenna 10 does not change, as in the absence. There is no significant change in characteristics. However, since the radio wave characteristics themselves change, it can be distinguished from absence and stationary. Even when a person is stationary, the characteristics vary depending on the stationary location and the stationary posture, so that the stationary position and the stationary posture can be identified.

  The transmitter Tx does not need a dedicated device specialized for the radio wave sensor device, and may be replaced with an existing transmitter such as a wireless LAN access point, ZigBee, or Bluetooth (registered trademark) module. .

  FIG. 2 is a functional block diagram showing the configuration of the main part of the space state determination apparatus 1 to which the present invention is applied. Here, the configuration unnecessary for the description of the present invention is omitted.

  The array antenna 10 is composed of a plurality of (channel) antenna elements at having directivity, and the antenna elements at are arranged in different directions. The feature amount calculation unit 11 includes a correlation matrix calculation unit 11a that calculates a correlation matrix from the vector information of the received wave, and calculates the feature amount of the received wave detected by each antenna element at. The spatial state estimation unit 12 estimates the state of the space at the time by applying the feature quantity of the received wave detected at each antenna element at each time to the learning model.

  The spatial state estimation unit 12 includes an SVM 12a as a learning model, represents received waves as multidimensional vectors having radio waves detected by the antenna elements at as features, and learned data obtained by classifying received waves according to spatial states. Then, the separation plane is learned, and the state of the space is estimated based on how the feature quantity calculated from the received wave is classified by the separation plane.

  The environmental sensor unit 13 includes environmental sensors such as a temperature sensor 13a, a humidity sensor 13b, and a CO2 sensor 13c, and repeatedly detects and outputs indoor temperature, humidity, CO2 concentration, and the like at a predetermined cycle. The behavior information collection unit 14 acquires behavior information such as a schedule registered by the user in advance in the cloud and a managed schedule, an on / off state of a home appliance such as a television or a lighting fixture, and outputs the behavior information together with current date and time information. Such collection of behavior information can be realized by accessing a pre-registered address on the cloud and acquiring schedule and smart home appliance on / off information.

  The spatial state estimation result, environmental sensor output, and behavior information are provided to the state determination unit 22 via the interfaces 15 and 21 and the network NW. The state determination unit 22 finally determines the state of the space at the time when the state determination is requested (state determination time) based on the estimation result of the spatial state, the environmental sensor output, and the behavior information.

  Next, a method of calculating the received wave feature value by the correlation matrix calculation unit 11a of the feature value calculation unit 11 will be described.

  When the antenna elements at are arrayed in a straight line in the array antenna 10, the reception signal of the array antenna 10 is a reception vector having the reception signal of each antenna element at as an element as shown in the following equation (1). x (t).

Where a (θ): L-dimensional vector when the number of antenna elements is L
s (t): Received signal at the reference point
n (t): Noise

  The L-dimensional vector a (θ) is expressed by the following equation (2).

Where θ is the direction of arrival of the radio wave relative to the direction of the array of antenna elements
d: Spacing between array elements
λ: Wavelength of radio wave

  Here, when M arriving waves arrive as plane waves, the following equation (3) holds.

  A is an L × M matrix having M vectors (referred to as steering vectors) as columns, and is represented by the following equation (4). S (t) is an M-dimensional vector having the complex amplitude of each incoming wave as an element, and is represented by the following equation (5). However, T represents transposition.

  The correlation matrix calculation unit 11a calculates a correlation matrix Rxx expressed by the following equation (6) from the reception vector x (t). However, the vector E represents a set average and H represents a complex conjugate transpose.

  Here, since the noise is unrelated to the incoming wave and independent of the element, the above equation (6) can be rewritten as the following equation (7). Here, σ is the variance of noise, and S is the wave source correlation matrix = E [s (t) s (t) H].

  Further, the correlation matrix Rxx can be expanded as shown in the following equation (9) using the eigenvalue λi obtained from the following equation (8) and the corresponding eigenvector vi.

here,

And diag is an array of diagonal elements of a matrix.

  The eigenvalues of the reception data correlation matrix Rxx can be divided into K signal eigenvalues corresponding to the sum of the number of coherent wave groups and incoherent waves, and L-K noise eigenvalues whose magnitude is equal to the noise power. That is, the following equation (11) is established. Here, diag is an array of diagonal elements of a matrix.

  From the above, it was shown that the correlation matrix Rxx generated from the received vector can be divided into a signal subspace and a noise subspace by expanding eigenvalues. The eigenvector v and the steering vector a (θ) spanning the signal subspace span the same space and can be expressed as the other linear combination. That is, the eigenvector spanning the signal subspace can be expressed by a linear combination of steering vectors including the direction of arrival information, and can be said to represent a radio wave propagation structure.

  The spatial state estimation unit 12 clusters the radio wave feature quantities using the SVM 12a. SVM is a kind of identification method using supervised learning, and learns parameters so as to obtain a separation plane having a maximum distance from given learning data. Basically, it is a technique for discriminating two classes, but two or more states can be discriminated by using a one-against-one method.

  SVM teacher data is constructed for each room and occupant to be monitored. In other words, if the SVM is used for two state identification, a separation plane is constructed in advance using the feature quantity of the received wave detected during absence and the feature quantity of the received wave detected when in the room as teacher data, The absence / occupancy is estimated based on which side of the separation plane the feature quantity of the received wave detected at time t is located.

  If the one-against-one method is used, teacher data is created for each location in the room and for each location (standing, lying, sitting, etc.). The separation plane is constructed in advance, and then the location and orientation are estimated depending on which side of the separation plane the feature quantity of the received wave detected at time t is located.

  The state determination unit 22 is notified from the discrete state estimation results periodically notified from the spatial state estimation unit 12, the time series of sensor outputs notified from the environment sensor unit 13, and the behavior information collection unit 14. The spatial state at the state determination time is determined based on the time series of the action information to be performed. For this determination, either (1) a rule-based state determination method or (2) a probability model-based state determination method can be applied.

  FIG. 3 is a functional block diagram illustrating the configuration of the state determination unit 22 when the rule-based state determination method is employed.

  In the history information database (DB) 131, history information of state determination results is stored together with the time information. FIG. 4 is a diagram showing an example of registration contents of the history information DB 131, in which history information of the state determination result by the state determination unit 22 is registered at a cycle of 1 second.

  The rule table 132 determines the state at the state determination time based on the state estimation result at each time and the state determination result and / or state estimation result within a relative determination period based on the time. Multiple rules are described. The rule search unit 133 searches the rule table 132 for a rule corresponding to the state estimation result at the state determination time. The contents of the rule table 132 can be viewed and edited by the rule browsing / editing unit 136.

  The history search unit 134 searches the history information DB 131 for the state determination result made in the determination period referred to by the searched rule. The rule base determination unit 130 determines the state at the time by applying the state determination result within the searched determination period to the rule searched based on the state estimation result at the state determination time. The history information registration unit 135 registers the state determination result as history information in the history information DB 131 together with the time information.

  5 and 6 are diagrams schematically representing the registered contents of the rule table 132. In the present embodiment, the rule table 132 includes a conditional expression table 132a and a determination rule table 132b. It is defined by the browsing / editing unit 136.

  In the conditional expression table 132a of FIG. 5, many conditions are registered with “relative time”, “conditional expression”, and “condition (value)” as items. “Relative time” represents the relative time relationship between the monitoring period of the spatial state and the current time, “NOW” represents the current time, “SINCE” represents the time before and after the time indicated, and “AFTER” Represents the time after that. Note that h after the number represents time, and m represents minutes. It may be expressed in seconds (s).

  The “conditional expression” is a specific condition and represents a condition relating to a spatial state estimation result by the spatial state estimation unit 12, a sensor output of the environment sensor 13, and behavior information. “Condition (value)” indicates the probability of time satisfying the conditional expression at the time in the case of% notation, and represents the numerical value in the case of a sensor value. The conditional expressions shown are as follows.

(1) Condition # 1
The current (NOW) estimation result is “0: absence” with a probability of 100%.

(2) Condition # 2
The current estimation result is “2: resident (stationary)” with a probability of 100%.

(3) Condition # 3
The judgment result from 10 minutes before to the present (SINCE-10m) is “0: absence” with a probability of 95% or more.

(4) Condition # 4
The determination result from 10 minutes ago to the present is “2: resident (stationary)” with a probability of 95% or more.

(5) Condition # 5
The average value of CO2 concentration from 5 minutes before to now (SINCE-5m) is more than 750ppm.

(6) Condition # 6
The power-on state of the air conditioner from 5 minutes ago to the present is on with 100% probability.

(7) Condition # 7
The estimated result from 5 minutes ago to the present is “1: occupancy (motion)” with a probability of less than 60%.

(8) Condition # 8
The schedule from the current time to 1 hour later is “meeting” with a probability of 100%.

  In the determination rule table 132b of FIG. 6, rules for performing state determination by combining the conditional expressions described in the conditional expression table 132a are described as a combination of “combined conditional expression” and “determination result”. ing. The numbers described in the “binding conditional expression” are the condition numbers (#) described in the conditional expression table 132a, and are expressed by AND, OR, and parentheses. Alternatively, it may be expressed as NOT, NOR, NAND, etc. When the combined conditional expression is satisfied, the determination result associated with the conditional expression is adopted. The illustrated example is as follows.

(1) Rule # 1
If the estimation result of the spatial state is “0: absence” or “2: resident (stationary)” and the judgment result from 10 minutes ago to the present is “0: absence” with a probability of 95% or more The space state is determined as “0: absence”. This is based on the fact that it cannot change from the absence state to the stationary state.

(2) Rule # 2
The estimation result of the spatial state is “0: absence” or “2: occupancy (stationary)” and the determination result from 10 minutes ago to the present is “2: occupancy (stationary)” with a probability of 95% or more. If the average value of the CO2 concentration from 5 minutes ago to the present is 750 ppm or more, the spatial state is determined as “2: resident (stationary)”. This is based on the fact that there is a person in the room when the occupancy is still in the room and the CO2 concentration is high, and it is unlikely that the person is away.

(3) Rule # 3
The estimation result of the spatial state is “0: absence” or “2: occupancy (stationary)” and the determination result from 10 minutes ago to the present is “2: occupancy (stationary)” with a probability of 95% or more. And the air-conditioning state of the air conditioner from 5 minutes ago to the present is on with a probability of 100%, the space state is determined as “2: resident (stationary)”. This is based on the fact that if the air conditioner is operating while still in the room until just before, there is a person in the room and it is unlikely that the person is away.

(4) Rule # 4
The estimation result of the space condition is “0: absence” or “2: resident (stationary)”, and the power-on state of the air conditioner from 5 minutes ago to the present is on with 100% probability, and 5 minutes If the determination result from previous to present (SINCE-5m) is “1: occupancy (motion)” with a probability of less than 60%, the spatial state is determined to be “2: occupancy (stationary)”. This means that if the air conditioner stays on for a while and the air conditioner is ON, even if it is estimated that the current time is absent and the occupant is still in the room, it is not going out but still in the occupant room. It is based on what is done.

(5) Rule # 5
If the estimation result of the spatial state is “0: absence” or “2: resident (stationary)” and the schedule from the current time to 1 hour later is “meeting” with a 100% probability, the spatial state Is judged as “0: absence”. This is based on the fact that if the meeting is scheduled, it is considered that the person is going out.

  According to the present embodiment, the state of the space is periodically estimated by applying the feature amount obtained from the received waves detected by the antenna array of the plurality of channels to the learning model, and further, the environment of the space and the human behavior Information is also detected, and the state of the space is finally determined from those time series. Therefore, if the time series of observed states does not match the typical state transition of the space or indicates a state transition that is not logically possible, the estimation by the learning model is incorrect. Therefore, it is possible to prevent state determination based on an erroneous estimation result and improve the accuracy of state determination.

  Further, according to the present embodiment, since the state of the space at the state determination time is determined based on the state estimation result at the time and the state observed within the determination period relatively determined from the time, State determination based on time series changes is facilitated.

  FIG. 7 is a functional block diagram showing the configuration of the state determination unit 22 that employs a probabilistic model-based determination method instead of the rule-based state determination method described above, and the same reference numerals as above denote the same or equivalent parts. Represents.

  The probability model base determination unit 137 includes a Bayesian network (BN) prepared in advance for the state estimation result at each time, the state determination result, the sensor output of the environment sensor 13 and the history information of the behavior information collected by the behavior information collection unit 14. ) To finally determine the state at the state determination time.

  The history information database (DB) 131 stores past state determination results together with the time information, and the contents thereof are the same as or equivalent to the contents described with reference to FIG. The history search unit 138 searches the history information DB 131 for a state determination result within the determination period requested from the probability model base determination unit 137 and responds. The history information registration unit 135 registers the state determination result as history information in the history information DB 131 together with the time information.

  BN is a model in which a qualitative dependency relationship of a random variable is represented by a graph structure, and a quantitative relationship between two random variables having a dependency relationship is represented by a conditional probability. When there are items that can be observed among the random variables of BN, it is possible to statistically infer variables that are not observed by using the dependency relationship between the random variables. Therefore, the state of the current space can be accurately determined by inputting the state estimation result by SVM, time information, and history information to the probability model base determination unit 137 having a determination function by BN.

  FIG. 8 is a diagram showing an example of the BN. In each node N, the posterior probability is learned in advance, the state estimation result by the spatial state estimation unit 12, and the state determination result stored in the history information DB The sensor output of the environmental sensor 13 and the behavior information collected by the behavior information collection unit 14 are taken as observed values (evidence variables). And the state of the space at the state determination time is determined probabilistically from these pieces of information. The definition of each state symbol is as follows.

Xt: Spatial state estimation result at time t calculated by the spatial state estimation unit 12
Sj, t: Sensor output of environment sensor j at time t
Ik, t: Acquisition value of action information at time t
S ^ j, t: Environmental judgment result calculated from sensor output of environmental sensor j from time t-1 to t
I ^ k, t: Action determination result calculated from the acquired value of action information from time t-1 to t
X ^ t: Judgment result of the spatial state at time t obtained probabilistically by BN from each state information

  In this embodiment, the environment determination result S ^ j, t is obtained based on the history information of the sensor output Sj, t of the environment sensor, and the action determination result I ^ is based on the history information of the acquired value Ik, t of the behavior information. k and t are obtained. Based on the estimation result Xt of the spatial state at the time t, the environmental judgment result S ^ j, t, the behavior judgment result I ^ k, t and the past spatial state judgment result (X ^ ti to X ^ t-1) Thus, the spatial state X ^ t at the time t is determined.

  FIG. 9 is a diagram showing an a posteriori probability table for obtaining the environment determination result S ^ j, t based on the sensor output Sj, t of the environment sensor, taking the case where the environment sensor 13 is the CO2 sensor 13c as an example. The posterior probability of the environment determination result S ^ j, t is determined based on the sensor output history (time t-1 to t). The specific value is determined by generating a random number. For example, if the CO2 concentration detected by the CO2 sensor 13c is 750 ppm or more, Sj, t = 1, and if it is less than Sj, t = 0, the condition based on the output of the CO2 sensor is that the person is in the room ( j, t = 1) or absent (S ^ j, t = 0).

  FIG. 10 is a diagram showing an example of a posterior probability table for obtaining the action determination result I ^ k, t based on the acquired value Ik, t of the action information, when the action is scheduled to be discussed. The posterior probability of the action determination result I ^ k, t is determined based on the information history (time t-1 to t). The specific value is determined by generating a random number. For example, if Ik, t = 1 if the schedule at time t is a meeting, and Ik, t = 0 if there is no schedule, then the status based on the schedule is I ^ k, t = 1, I It is determined by the posterior probability of ^ k, t = 0. Note that it is not efficient to store all posterior probabilities as the number of states increases, and therefore, the posterior probabilities are calculated based on the posterior probabilities of each information source and conditional probabilities P (I ^ k, t | Ik, t).

  FIG. 11 shows the state estimation result Xt at the state determination time t, the environment determination result S ^ j, t, the action determination result I ^ k, t, and the past spatial state determination results (X ^ ti to X ^ t-1). 6 is a diagram showing an example of a posterior probability table for obtaining a spatial state determination result X ^ t at a state determination time t based on FIG. In addition, since it is not efficient to memorize all the posterior probabilities when the number of states increases, the estimation result of the spatial state, the posterior probability of the environment judgment result S ^ j, t, and the action judgment result I ^ k, t Posterior probabilities, posterior probabilities of past spatial state judgment results (X ^ ti to X ^ t-1), and conditional probabilities P (X ^ t | Xt), P (X ^ t | S ^ j, t), P (X ^ t | I ^ k, t), P (X ^ t | X ^ ti).

  In the above embodiment, the spatial state is finally determined based on the spatial state estimation result, the environmental sensor output, and the behavior information. However, the present invention is not limited to this. When the behavior information cannot be acquired, the spatial state may be finally determined based on the spatial state estimation result and the environmental sensor output.

  DESCRIPTION OF SYMBOLS 10 ... Array antenna, 11 ... Feature-value calculation part, 11a ... Correlation matrix calculation part, 12 ... Spatial state estimation part, 13 ... Environmental sensor part, 22 ... State determination part, 130 ... Rule base determination part, 131 ... History information database , 132 ... rule table, 132a ... conditional expression table, 132b ... determination rule table, 133 ... rule search unit, 134, 138 ... history search unit, 135 ... history information registration unit, 136 ... rule browsing / editing unit, 137 ... probability Model-based judgment unit

Claims (7)

In a spatial state determination device that receives radio waves transmitted from a transmitter with an antenna array of a plurality of channels and determines a state of a space in which the radio waves propagate based on received waves of each antenna array,
A feature amount calculating means for periodically calculating a feature amount of a received wave detected by each antenna array;
Spatial state estimation means for estimating the state of the space by applying the feature quantity of each period to a predetermined learning model;
An environmental sensor for detecting the environment of the space;
A spatial state determination device comprising: a spatial state determination unit that determines a spatial state at a predetermined state determination time based on a spatial state estimation result at the time and an output of an environmental sensor. Further comprising behavior information obtaining means for obtaining behavior information of a person acting in the space;
The said space state determination means determines the state of the space at a predetermined state determination time based on the estimation result of the space state at the time, the output of the environmental sensor, and the behavior information. Spatial state determination device.   The spatial state estimation means includes an SVM as a learning model, represents a received wave as a multidimensional vector having radio waves detected by each antenna array as a feature, and separates the received wave from learning data classified according to a spatial state. The space state according to claim 1 or 2, wherein the state of the space is estimated based on how the feature quantity calculated from the received wave is classified by the separation plane. State determination device.   The space state determination means determines the state of the space at the state determination time based on a state estimation result at the time and a state observed within a predetermined determination period relatively determined from the time. The space state determination apparatus according to claim 1.   The space state determination means determines the state of the space at the state determination time based on a state estimation result at the time and a probability that the predetermined state is observed within a predetermined determination period relatively determined from the time. The space state determination apparatus according to claim 4, wherein The space state determination means includes
History information DB that accumulates history information of state determination results together with time information;
A rule table in which a plurality of rules for determining the state based on the state estimation result and the history information of the state determination result are described;
Rule search means for searching a rule corresponding to the state estimation result of the state determination time from the rule table;
History information search means for searching the history information DB for a state determination result within a predetermined determination period referred to by the searched rule;
The space state determination device according to claim 4, further comprising a rule base determination unit that applies a state determination result within the determination period to the retrieved rule to determine a state at the state determination time. . The state determination means includes
History information DB that accumulates history information of state determination results together with time information;
Probability model-based determination means for finally determining the state at the state determination time by inputting history information of the state estimation result and the state determination result to a Bayesian network prepared in advance,
A history search unit that searches and responds to the state determination result within the determination period requested from the probability model base determination unit from the history information DB;
6. The space state determination apparatus according to claim 5, further comprising history information registration means for registering the state determination result as history information together with the time information in the history information DB.