JP2019082363A - Device for estimating terminal location on basis of presence probability density in consideration for radio wave attenuation, program therefor and method therefor - Google Patents

Abstract

To provide a device that can more accurately estimate a location of a terminal receiving a radio wave from a radio wave source even in a situation where a movable object can exist.SOLUTION: The present device is a terminal location estimation device that acquires reception radio wave information in a terminal to estimate a location of the terminal, and the terminal location estimation device has: object distribution determination means that acquires object recognition information from object recognition means capable of recognizing a movable object, and determining an object distribution on the basis of the object recognition information and arrangement information on an object installed beforehand; radio wave attenuation information determination means that determines radio wave attenuation information pertaining to a level of radio wave attenuation due to the object in between a radio wave source and a location in the location estimation area on the basis of the determined object distribution in the location estimation area; and terminal location determination means that determines a location of the terminal on the basis of an existence probability density function about the location in the location estimation area and denoting an amount pertaining to a received radio wave reflecting the determined radio wave attenuation information as a variable.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a technique for estimating the position of a terminal based on the reception status of radio waves at the terminal.

  Conventionally, techniques for estimating the position of a receiving terminal using a plurality of received radio signals (radio waves) have become widespread. For example, positioning techniques using GPS (Global Positioning System) outdoors, wireless LAN (Local Area Network) indoors, etc. are widely used at present.

  In the positioning technology using such radio waves, whether or not there is an object that affects the propagation of radio waves between a radio wave source (radio wave source) such as a satellite, a base station or an access point, and a terminal, It greatly influences the positioning result. For example, when there is a shield that blocks radio waves, the received signal strength (RSS, Received Signal Strength) at the terminal is lower than when radio waves reach the terminal directly, and the distance between the radio source and the terminal is greater than actual By making a large estimate, the error in position estimation will increase.

  For such problems, the system disclosed in Non-Patent Document 1 is different from the radio wave attenuation model used in the case where there is a static shield with a fixed position between the transmitting and receiving nodes. By using the model, the position estimation accuracy is improved. In addition, it estimates the error of position estimation by radio wave simulation taking into consideration the influence of the shield, and determines the position of the transmitting node so that the predicted error becomes smaller, so that the optimal transmission node arrangement according to the shield is obtained. You can get it.

  Furthermore, in Patent Document 1, when there is an obstacle of radio wave transmission between radio wave transceivers, the radio signal between the radio emission and reception unit is not derived from the radio characteristic parameter between the radio emission and reception unit and the radio terminal. A technique is disclosed for detecting the obstacle present between radio waves from changes in the radio wave propagation environment by communication and estimating the position of a wireless terminal. In this technology, when radio waves are blocked, priority (likelihood) to be used for position estimation is set for each transmitter.

JP, 2011-203129, A

Oka Kaito, Masaki Igarashi, Hideaki Uchiyama, Keiji Shimada, Ichi Nagahara, Rinichiro Taniguchi, "Anchor Node Placement Considering Occlusion in Wireless Position Estimation", Multimedia, Distributed, Cooperative and Mobile (DICOMO 2015) Symposium, 2015 , 799-802.

  In the prior art including the technique described above, although the object affecting the propagation of the radio wave is considered, the considered object is only a static shield or obstacle fixed in position. . However, in the actual site of terminal position estimation, there are many cases where a person can not be ignored as an object that interferes with radio wave (radio signal) propagation.

  For example, currently, in stores of various fields, there is an increasing need for accurately grasping the flow lines of a customer and a clerk who owns a terminal. In this case, it is general to estimate a terminal position by arranging a plurality of beacon transmission sources (hereinafter referred to as beacons) in a store. Here, the person moving in the store naturally has a very large influence on the terminal position estimation by these beacons.

  However, in the prior art including the above-described technique, such a person can not be considered as an object that affects the propagation of radio waves. As a result, an increase in error in the terminal position estimation result becomes a problem. As described above, although the influence of the movable object including the person on the radio wave propagation is great, it has not been taken into consideration in the past.

  Therefore, the object of the present invention is to provide an apparatus, a program and a method capable of more accurately estimating the position of a terminal that has received a radio wave from a radio wave source even in a situation where a movable object may exist. Do.

According to the present invention, there is provided a terminal position estimation apparatus for acquiring received radio wave information in a terminal that has received radio waves from a radio wave source and estimating the position of the terminal,
The object recognition information from the object recognition means capable of recognizing the movable object is acquired, and the movable object and the previously installed object are obtained based on the object recognition information and the arrangement information of the previously installed object Object distribution determining means for determining an object distribution which is a distribution of objects;
Radio wave attenuation information determination means for determining radio wave attenuation information related to the degree of radio wave attenuation by the object between the radio wave source and the position in the position estimation region based on the object distribution determined in the position estimation region ,
The position of the terminal is determined based on the existence probability density function which is a presence probability density function for the position in the position estimation area and the amount of the received radio wave reflecting the determined radio wave attenuation information as a variable A terminal position estimation device is provided which comprises:

  As one embodiment of the terminal position estimation apparatus according to the present invention, the above-mentioned "amount related to received radio wave reflecting radio wave attenuation information" means the number of the objects existing between the radio wave source and the position and each object It is also preferable that the received radio wave intensity that reflects the radio wave attenuation by the object, which is calculated from an amount representing the degree of radio wave attenuation due to and the reference received radio wave intensity as a reference.

  Further, in the above embodiment, the terminal position determination means adopts at least a corresponding predetermined amount for each of the movable object and the previously installed object as an amount indicating the degree of the radio wave attenuation. Is also preferred.

  Furthermore, the object distribution determination means according to the present invention may set a predetermined position range including the estimated location of the movable object estimated from the object identification information as the distribution range of the movable object. preferable.

Further, as another embodiment of the terminal position estimation device according to the present invention, the radio wave source is a plurality of radio wave sources arranged in advance,
The terminal position determination means selects a suitable radio wave source suitable for position estimation of the terminal from the plurality of radio wave sources based on the existence probability density function, and is selected among the acquired received radio wave information It is also preferable to determine the position of the terminal using information on a plurality of the preferred radio wave sources.

  Furthermore, in the other embodiment described above, the terminal position determination means is configured to receive the plurality of the suitable radio wave sources based on the presence probability density function of the plurality of the selected radio wave sources selected. It is also preferable to determine the position of the terminal using radio wave information. Here, it is also preferable that a plurality of the suitable radio wave sources are further selected for a predetermined number in order from the highest radio wave intensity among the suitable radio wave sources.

  In the other embodiment described above, the terminal position determination means does not change the object distribution at one time for position estimation if the object distribution at one time does not change by a predetermined amount or more as compared with the object distribution at the previous time. It is also preferred that the preferred radio source selected at the previous time be the preferred radio source at one of the above times.

  Furthermore, in the other embodiment described above, the terminal position determination means is configured to transmit all of the plurality of arranged radio wave sources to the preferred radio wave at one time described above when there is no movable object at one time related to position estimation. It is also preferable to use it as a source.

  Further, in the terminal position estimation device according to the present invention, the object recognition information may be taken from at least one of a camera, an omnidirectional camera, a depth camera, an infrared camera, an infrared positioning means, a laser positioning means, and a thermographic means. It is also preferable to generate based on measurement information.

  Furthermore, as still another embodiment of the terminal position estimation device according to the present invention, the error of the estimated position of the terminal is minimized based on the existence probability density function at least with respect to the object arrangement according to the arrangement information of the object previously installed. The radio wave source arranging means further determines the arrangement position of the radio wave sources to be arranged in advance from the candidate positions where the radio wave sources can be arranged based on the arrangement determined. It is also preferable to have.

According to the present invention, there is also provided a program that causes a computer mounted on a device to obtain received radio wave information at a terminal that has received radio waves from a radio wave source and to estimate the position of the terminal.
The object recognition information from the object recognition means capable of recognizing the movable object is acquired, and the movable object and the previously installed object are obtained based on the object recognition information and the arrangement information of the previously installed object Object distribution determining means for determining an object distribution which is a distribution of objects;
Radio wave attenuation information determination means for determining radio wave attenuation information related to the degree of radio wave attenuation by the object between the radio wave source and the position in the position estimation region based on the object distribution determined in the position estimation region ,
The position probability of the terminal is determined based on the presence probability density function with the amount of the received radio wave information reflecting the determined radio wave attenuation information as a variable. A terminal position estimation program is provided which causes a computer to function as terminal position determination means to determine.

According to the present invention, there is further provided a terminal position estimation method by a computer mounted on an apparatus for acquiring received radio wave information in a terminal that has received radio waves from a radio wave source and estimating the position of the terminal.
The object recognition information from the object recognition means capable of recognizing the movable object is acquired, and the movable object and the previously installed object are obtained based on the object recognition information and the arrangement information of the previously installed object Determining an object distribution, which is a distribution of objects;
Determining radio wave attenuation information according to the degree of radio wave attenuation by the object between the radio wave source and the position in the position estimation region based on the object distribution determined in the position estimation region;
The position probability of the terminal is determined based on the presence probability density function with the amount of the received radio wave information reflecting the determined radio wave attenuation information as a variable. A terminal position estimation method is provided comprising the steps of:

  According to the terminal position estimation apparatus, program, and method of the present invention, it is possible to more accurately estimate the position of a terminal that has received a radio wave from a radio wave source, even in a situation where a movable object may exist.

FIG. 1 is a schematic view showing an embodiment of a terminal position estimation system including a terminal position estimation device according to the present invention. It is a functional block diagram which shows the function structure in one Embodiment of the terminal position estimation apparatus by this invention. It is a distribution map of existence probability density computed by the electric wave from the beacon provided in the position presumption field. It is a schematic diagram for demonstrating the number of objects as electromagnetic wave attenuation information, and the electromagnetic wave attenuation amount of each object. It is a flowchart which shows one Embodiment of the terminal position estimation process in an object recognition information management part, an object distribution determination part, an electromagnetic wave attenuation information determination part, and a terminal position determination part. It is a flowchart which shows one Embodiment of the suitable electromagnetic wave source selection process in an electromagnetic wave source selection part.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[Terminal position estimation system]
FIG. 1 is a schematic view showing an embodiment of a terminal position estimation system including a terminal position estimation device according to the present invention.

  The terminal position estimation system according to the present embodiment shown in FIG. 1 estimates the position every moment in one store in at least one terminal 3 possessed by a store clerk, a shopper or the like, and possesses the terminal 3 It is possible to grasp the flow line in the store such as the store clerk or the customer.

Specifically, this terminal position estimation system
(A) Beacons 2 installed at multiple locations in a store and transmitting radio waves at predetermined time intervals in all directions;
(B) one or more cameras 4 capable of shooting movable objects such as people and transmitting information of the captured image in time series via a wireless or wired communication network;
(C) “Object recognition information” is generated from time-series images acquired from the camera 4 via the communication network, and further directly from the terminal 3 via the wireless communication network (or via a relay device such as an access point) ) A terminal position estimation device 1 for acquiring "received radio wave information", estimating the position of the terminal 3 every moment based on the generated / acquired information, and determining the flow line in the store.

  Here, in the store which is a position estimation area, an object that affects radio wave propagation such as a desk, a chair, a partition, a display stand, and a counter is disposed. The terminal position estimation device 1 also acquires such arrangement information (“object arrangement information”) of an object previously installed in the shop and uses it for the position estimation of the terminal 3. Note that this “object arrangement information” may also be generated based on the image group from the camera 4 after recognizing the installed object. As a result, even if, for example, the layout of the equipment in the store is changed or there is a temporary placement, it is possible to obtain “object arrangement information” that appropriately corresponds to it.

  Also, the beacon 2 can be, for example, a beacon of Bluetooth® Low Energy (BLE) specification that transmits advertisement packets in all directions at intervals of 100 to 1000 msec. Naturally, the beacon 2 is not limited to this. For example, an access point of a wireless LAN such as Wi-Fi (registered trademark) can be used, and in addition, the terminal 3 can receive radio waves from there and generate "received radio wave information". Then, various radio wave sources can be adopted as the beacon 2.

  Incidentally, the terminal position estimation device 1 determines the installation position of the beacon 2 as shown in FIG. 1 (in-store arrangement of the beacon 2) using captured image data from the camera 4 as will be described in detail later. It is also good.

  Furthermore, in the present embodiment, the terminal 3 is a smart phone, a mobile phone, a tablet computer, a wearable terminal, etc. that can be carried by a store clerk, a shopper, etc. Applicable The terminal 3 can simultaneously receive radio waves transmitted from a plurality of (preferably three or more) beacons 2.

Similarly, the terminal position estimation device 1 shown in FIG. 1 is a device that acquires “received radio wave information” in the terminal 3 that has received radio waves from the beacon 2 that is a radio wave source, and estimates the position of the terminal 3 Specifically,
(A) Acquire "object recognition information" from object recognition means capable of recognizing movable objects, and move based on this "object recognition information" and "object arrangement information" of an object installed in advance An object distribution determination unit 113 that determines an “object distribution” that is a distribution of objects that can be set and objects that are installed in advance;
(B) A radio wave by the object between the beacon 2 as a radio wave source and the position in the position estimation area (store) based on the "object distribution" determined in the position estimation area (store in this embodiment) A radio wave attenuation information determination unit 114 that determines "radio wave attenuation information" related to the degree of attenuation;
(C) The "existence probability density function" which is a "presence probability density function" about the position in the position estimation area (store) and which has the amount of the received radio wave reflecting the determined "radio wave attenuation information" as a variable And a terminal position determination unit 116 that determines the position of the terminal.

  As described above, the terminal position estimation device 1 uses the "presence probability density function" having a variable reflecting the "radio wave attenuation information" determined based on the "object distribution" which can include a movable object such as a person or the like. , Determine the terminal position. As a result, even in the situation where movable objects may exist, it is possible to consider how much the received radio wave is attenuated by the “object distribution that may include movable objects”, and the radio waves from the radio wave source The position of the received terminal can be estimated more accurately.

  Here, the object recognition means of the above configuration (A) may be a functional configuration part in the present device 1 as shown as an object recognition part 111 in FIG. 2 later, or provided in an external device It may be In any case, it is also preferable that the object recognition means can perform tracking by giving a video flow line ID (identifier) to a movable object recognized in the image.

  Even in such a case, the terminal position estimation device 1 does not take into account which video flow line ID is associated with which position estimation target terminal, and "presence probability density reflecting the" object distribution " The function can be used to determine the position of this terminal. That is, it is possible to carry out terminal position estimation processing more simply without the need for matching technology between terminal ID and video flow line ID, and at the same time, there is a problem that reliability of presence probability density in maximum likelihood estimation is lowered. It can also be solved.

  As a modification of the object recognition unit 111 (FIG. 2) that generates the "object recognition information" based on the photographed image data from the camera 4, an object recognition unit (object recognition unit 111) is provided in an external device. If it is, the terminal position estimation device 1 obtains "object recognition information" from this device.

  Further, in the terminal position estimation device 1, for example, the “object distribution” is determined based on the “object recognition information” and the “object arrangement information” at each time of the position estimation period, and the position of the terminal 3 is estimated every moment. It also becomes possible. As a result, even under the situation where, for example, a large number of persons move around the store and affect the radio waves from the beacon 2, the flow line of each terminal 3 can be determined with higher accuracy.

  Incidentally, the above-mentioned "received radio wave information" is, for example, information in which the reception time and received signal strength (Received Signal Strength Indicator, RSSI) in the radio wave received by the terminal 3 from the beacon 2 are associated with the ID of the beacon 2. be able to. In addition, “object arrangement information” may be, for example, position coordinates or position coordinate range of an object arranged in advance. Furthermore, “object recognition information” can be, for example, the foot position coordinates of a recognized person at each time.

  Further, as another embodiment, the “object recognition information” may be an image taken from at least one of an omnidirectional (all direction) camera, a depth camera, an infrared camera, an infrared positioning means, a laser positioning means, and a thermographic means It may be generated based on measurement information. For example, depth value information (of each target pixel) generated by the depth camera may be converted into image data, and may be used in the same manner as an image captured by the camera 4. In any case, the terminal position estimation device 1 can acquire the "object recognition information" generated by some means and use it for position estimation of the terminal 3.

  Further, the movable object as the recognition target according to the "object recognition information" may be a person, an animal, a vehicle, or any other movable physical object such as a movable physical object if it can be recognized by the object recognition means (for example If possible) various things apply. Also, the place to be photographed is not particularly limited, and it may be indoors such as a company, a school, a hall or the inside of a house other than in a store, for example, and also a spectator or a commuter , Shoppers, workers, pedestrians, runners, etc. may be visible outdoors.

  Furthermore, as still another embodiment, the terminal 3 includes the above configurations (A) to (C), and as a terminal position estimation device that estimates its own position and the positions of other terminals 3 in the vicinity. It may function. In this case, it is also possible to generate “object recognition information” using, for example, a camera incorporated in the terminal 3 (the position of which has been determined).

[Configuration / Function of Terminal Position Estimation Device]
FIG. 2 is a functional block diagram showing a functional configuration in an embodiment of the terminal position estimation apparatus according to the present invention.

  According to FIG. 2, the terminal position estimation device 1 has a communication interface 101 capable of communication connection with the camera 4, the terminal 3, etc., a received radio wave information storage unit 102, an object recognition information storage unit 103, and an object arrangement information storage unit. And 104, a terminal flow line information storage unit 105, a display / keyboard (DP · KB) 106, and a processor / memory. Here, the processor memory stores one embodiment of the terminal position estimation program according to the present invention, and has a computer function, and executes the terminal position estimation program to perform terminal position estimation processing. Conduct.

  Furthermore, the processor memory includes an object recognition unit 111, an object recognition information management unit 112, an object distribution determination unit 113, a radio wave attenuation information determination unit 114, a radio wave source placement unit 115, and a radio wave source as functional components. A terminal position determination unit 116 including a selection unit 116a, a terminal flow line information management unit 117, a communication control unit 121, a received radio wave information management unit 122, and an input / output control unit 123 are included. These functional components can be regarded as the functions of the terminal position estimation program stored in the processor memory, and the functional components of the terminal position estimation apparatus 1 in FIG. The processing flow is also understood as one embodiment of the terminal position estimation method according to the present invention.

  Similarly, in FIG. 2, the camera 4 is an imaging device compatible with visible light, near infrared rays, or infrared rays provided with a solid-state imaging device such as, for example, a CCD image sensor or a CMOS image sensor. In addition, the camera 4 or the camera control device (not shown) generates photographed image data including an image of an object photographed by the camera 4 and transmits the information to the terminal position estimation device 1 in time series or batch. Have a function to

  The communication interface 101 receives, from the camera 4 or the camera control device, photographed image data which is a time-series image group through a wired or wireless communication network. The acquired captured image data (image file) is output to the object recognition unit 111. The communication interface 101 also receives received radio wave information from the terminal 3 directly (or via a relay device such as an access point) via a wireless communication network. Alternatively, the reception radio wave information may be received after being wired connected to the terminal 3 via a USB (Universal Serial Bus) or the like.

  The communication interface 101 may further transmit flow line information generated as a result of terminal position estimation to an external information processing apparatus. The transmission and reception of the information in the communication interface 101 is controlled by the communication control unit 121, and the information is transmitted and received at an appropriate timing.

  In the present embodiment, the object recognition unit 111 recognizes a person who stays or moves in the store, which is a position estimation area, based on photographed image data related to an image photographed by the camera 4 and generates object (person) recognition information Do. Here, even if the generated object (person) recognition information includes two-dimensional (or three-dimensional) coordinate values of the foot position in one or a plurality of recognized people at each time of recognition processing. Good. The object (person) recognition process in the object recognition unit 111 can be implemented by adopting various known methods.

  For example, when using an omnidirectional camera as the camera 4, non-patent documents Tatsuya Kobayashi, Haruhisa Kato, Masaru Ogino, "Proposal of a person flow line tracking method that is robust against shielding using a monocular omnidirectional camera", Object (person) recognition information can be generated using the method described in the technical report, vol. 116, no. 411, PRMU 2016-146, 2017, pp. 321-326.

  Specifically, in this method, a person detection process and a person tracking process are performed for each image frame, the position of each person at each time is estimated, and the person is tracked. Among them, in the human detection process, the region extracted using the background subtraction technique is matched with the 3D human model prepared in advance, and the human region in the image is detected. Here, a human model is arranged in advance at various foot positions where a human can exist, and a corresponding foot position is detected as a human position by searching for an outline having the highest similarity to the foreground area.

  Furthermore, in the person tracking process, the person ID being tracked is assigned to the detected person area by associating the person being tracked with the above detection result, and the element whose matching score becomes maximum by one-to-one person matching is selected. select. Then, based on the result of the person matching, a particle filter is used to estimate the tracking position of each person.

  Incidentally, the object recognition unit 111 is not provided in the present device 1, and the object recognition information management unit 112 can take an embodiment in which object recognition information is acquired and managed from an external object recognition unit. In any case, the object recognition information management unit 112 appropriately outputs the acquired object recognition information to the object distribution determination unit 113 while storing the acquired object recognition information in the object identification information storage unit 103.

  The object distribution determination unit 113, based on the input object recognition information and the input object arrangement information, position distribution of an object (for example, a person) that can move and a previously installed object (for example, facilities in a store) Determine the object distribution

  Here, the object arrangement information is information including the coordinate value or coordinate value range of the arrangement position of the object affecting the radio wave propagation such as desk, chair, partition, display stand, and counter in the store, from the keyboard 106 It is also preferable that the information is input via the input / output control unit 123 or from an external information processing apparatus or the like via the communication interface 101, recorded in the object arrangement information storage unit 104, and output to the object distribution determination unit 113. It is also preferable that the object arrangement information further includes information on the (longitudinal and lateral) size of the in-store floor on which position estimation is performed. Note that, as a modification, this object arrangement information may also be generated by the object recognition unit 111 recognizing an installed object in a store based on captured image data related to an image captured by the camera 4.

Although described later with reference to FIG. 4, the object distribution determination unit 113 moves a predetermined position range including an estimated location of a movable object (for example, a person) estimated from object identification information. It is also preferable to use the distribution range of the object to be obtained. Here, this predetermined position range may be, for example, a range of a circle of radius c (m) centered on the estimated location (x _p , y _p ).

The radio wave attenuation information determination unit 114 determines, based on the object distribution in the position estimation area (store) determined by the object distribution determination unit 113, the beacon 2 and the position (x, y) in the position estimation area (store). The radio wave attenuation information related to the degree of radio wave attenuation by the object during As this radio wave attenuation information,
(A) Number of objects present between beacon 2 and position (x, y), and (b) radio wave attenuation by each object present r _loss (m)
Can be determined. Of course, the radio wave attenuation information is not limited to the above (a) and (b), and various information as long as it is information of an amount related to the received radio wave (for example, RSSI) as the degree of radio wave attenuation. Can be adopted.

Here, the radio wave attenuation amount r _loss (m) in the above (b) indicates an attenuation amount equivalent to the degree of attenuation while the radio wave propagates in vacuum by the distance r _loss (m). An amount preset in advance may be used as the radio wave attenuation amount r _loss , but for example, an object that can move (a person in the present embodiment) and an object installed in advance (a facility in a store, etc. in the present embodiment) And may set different amounts adapted to each. Furthermore, if information on the material of each object (for example, a metal material such as a human body, a wall material, a desk, etc.) can be obtained, it is also preferable to set the radio wave attenuation amount suitable for the material for each material.

  The radio wave source placement unit 115 determines, from among candidate positions where the beacon 2 can be placed in the position estimation area (store), a placement position where the beacon 2 is actually placed in advance. The arrangement position determination process of the beacon 2 in advance will be described in detail later, but at least the object arrangement related to the arrangement information of the object installed in advance is based on the existence probability density function described in detail later. The arrangement of beacons 2 for minimizing the error of the estimated position is determined, and based on the determined arrangement, the arrangement positions at which a plurality of beacons 2 are previously arranged are determined from the candidate positions where beacon 2 can be arranged. It may be one.

  Note that, as a modification, the device 1 does not have the radio wave source placement unit 115, and the placement information of the beacon 2 that is deemed appropriate from the sense of experience or in the past is input / output control from the keyboard 106 The information may be input to the terminal position determination unit 116 via the communication interface 101 via the unit 123 or from an external information processing apparatus or the like.

  The received radio wave information management unit 122 receives and manages the received radio wave information acquired from the terminal 3 while storing the received radio wave information in the received radio wave information storage unit 102, and appropriately outputs the information to the terminal position determination unit 116. Here, the received radio wave information can be information in which the reception time, the RSSI, and the ID of the beacon 2 are associated with each beacon 2 that is the transmission source of the radio wave received by the terminal 3.

  Similarly, in FIG. 2, the terminal position determination unit 116 is an existing probability density function p (rssi | pos) for the position pos in the position estimation area (store), and the received radio wave reflecting the radio wave attenuation information determined. The position of the terminal 3 is determined based on the existing probability density function p (rssi | pos) with the amount rssi as a variable.

Here, the variable “amount rssi related to received radio wave reflecting radio wave attenuation information” is
(A) The number num of objects existing between the beacon 2 and the position pos, and an amount indicating the degree of radio wave attenuation by each object, for example, the radio wave attenuation amount r _loss described above
(B) It can be set as the RSSI reflecting the radio wave attenuation by the object, which is calculated from the reference received radio wave intensity rssi _{0 as the} reference. Hereinafter, one specific example of this variable and the existing probability density function p (rssi | pos) will be described.

In general, there are n beacons 2 and the terminal 3 exists at a position pos at a distance of r _{pos, i} (m) to the i-th (i = 1, 2,..., N) beacon 2 of them. The received signal strength (RSSI) r ssi _{i at} is given by the following equation (1) rssi _i = rssi ₀ −10 d · log ₁₀ (r _{pos, i} )
Is approximately represented. Here, rssi ₀ is the RSSI measured at a position 1 m away from the beacon 2, and d is a constant representing the degree of attenuation.

Also, assuming that the existence probability density follows a normal distribution, the existence probability density function p (rssi _i | pos) that receives radio waves of intensity rssi _i at this position pos is expressed by the following equation (2) p (rssi _i | pos ) = (2πσ ² ) ^-0.5 · exp {-(rssi _i -A _{pos, i} ) ² / 2σ ² }
Is represented by Here, σ ² is the variance of the RSSI, and A _{pos, i} is an ideal reference (for example, in a vacuum without any object) of the radio wave from the i-th beacon 2 at position pos It is RSSI.

  Next, consider the case where an object that reflects or shields radio waves is present in the position estimation area (store). The radio wave reaching the terminal 3 at the position pos is attenuated due to the effect of the object existing between it and the beacon 2, so the distance value from the beacon 2 calculated simply from this RSSI is greater than the actual distance. As a result, the position estimation accuracy is reduced.

  FIG. 3 is a distribution diagram of the presence probability density calculated by radio waves from the beacon 2 provided in the position estimation area.

In FIG. 3, the distribution of the existing probability density p indicated by shading is under the condition that the static installation is not installed on the floor in the store (position estimation area). In this case, the position (dark gray portion) where the probability of the presence of the terminal 3 (the user who has the terminal 3) is high is on the circumference of the radius r _{rssi = Apos} centering on the beacon 2 according to the above equation (2). On the other hand, in the case where a static installation as shown in FIG. 3 is installed, the position where the terminal 3 is considered to be present (high presence probability density) is actually a part of the circle squeezed It becomes on the periphery of such a shape (shape shown by the thick line in FIG. 3).

Therefore, radio wave attenuation information pertaining to the object is taken in advance in rssi _i which is a variable of the existence probability density function, and the existence probability density is handled with a variable reflecting the influence of radio wave attenuation by the object. This enables more accurate position estimation in consideration of the influence of existing objects. Incidentally, since the influence of the object at each time is taken in advance as a variable into the calculation of the existence probability density, it is possible to easily handle a movable object such as a person as an object reflecting the influence.

Here, as described above as the radio wave attenuation information, using the number num of objects existing between the beacon 2 and the position pos and the radio wave attenuation amount r _loss of each object, the received signal strength (RSSI) of the position pos rssi _i is a modification of the above equation (1), and the following equation (3) rssi _i = rssi ₀ −10 d · log ₁₀ (r _{pos, i} + num · r _loss )
It is expressed as In the form of (r _{pos, i} + num · r _loss ) in the above equation (3), radio waves by an object existing between beacon 2 and position pos whose distance between them is r _{pos, i} (m) It is understood that the RSSI originally obtained at the position pos can be obtained at a position where the distance from the beacon 2 is r _{pos, i} + r _loss (m).

Thereby, the existence probability density function p (rssi _i | pos) for receiving radio waves of intensity rssi _i at the position pos is expressed by the following equation (4) p by substituting the above equation (3) into the above equation (2) (rssi _i | pos) = (2πσ ² ) ^−0.5 · exp (−B)
B = (rssi ₀ −A _{pos, i} −10 d · log ₁₀ (r _{pos, i} + num · r _loss )) ² / 2σ ²
It is expressed as

By the way, in the above process, the influence of the radio wave attenuation by the object is taken into the existence probability density function by reflecting the radio wave attenuation information to the RSSI as in the above equation (3). The radio wave attenuation information may be reflected on the ideal RSSI A _{pos, i in the above} . Specifically, A _{pos, i} is expressed by the following equation (5): A _{pos, i} = rssi ₀ −10 d · log ₁₀ (r _{pos, i} + num · r _loss )
It can be expressed as

Further, the existence probability density function p (rssi _i | pos) in this case can be calculated by substituting the above equation (5) into the above equation (2) to obtain the following equation (6) p (rssi _i | pos) = (2πσ ² ) ^-0.5 · exp (-C)
_{C = (rssi 0 -rssi i -10d} · log 10 (r pos, i + num · r loss)) 2 / 2σ 2
It is expressed as

As described above, either the above equation (4) or the above equation (6) can be adopted as the existing probability density function p (rssi _i | pos) based on a variable taking into consideration the object distribution. In any case, the variables of the existing probability density function are the RSSI reflecting the radio wave attenuation by the object, which is calculated from num and r _loss and the reference received radio wave intensity as a reference.

Further, consider a situation where the terminal 3 present at the position pos receives radio waves simultaneously from each of the n beacons 2. Assuming that a set of RSSI is RSSI = {rssi ₁ , rssi ₂ ,..., Rssi _n }, the probability of presence probability density function p (RSSI | pos) by all received radio waves is independent of the probability of presence probability of each beacon 2 Therefore, in the end, the following equation (7) p (RSSI | pos) = p (rssi ₁ , rssi ₂ , ..., rssi _n | pos)
= P (rssi ₁ | pos) p (rssi ₂ | pos) ... p (rssi _n | pos)
Will be represented by

Here, in the above equations (3) to (6), it is assumed that the radio wave attenuation amount r _loss of each object is a predetermined single value, but a different radio wave attenuation amount is acquired for each object type It is also preferred to use those values. For example, the radio wave attenuation amount r _{loss, 1} (m) and the number of persons num ₁ (person) of a person and the radio wave attenuation amount r _{loss, 2} (m) of the in-store equipment installed in advance and the in-store equipment number num ₂ ( And the part of r _{pos, i} + num · r _{loss in} the above equations (3) to (6),
(8) r _{pos, i} + num ₁ · r _{loss, 1} + num ₂ · r _{loss, 2}
It is also preferable to use in place of Hereinafter, the situation of the above equation (8) will be described with reference to FIG.

  FIG. 4 is a schematic view for explaining the number of objects as radio wave attenuation information and the radio wave attenuation amount of each object.

In the example shown in FIG. 4, there is one person as a movable object between the beacon 2 and the position pos, and two in-store equipment as a pre-installed object. There is. Therefore, num ₁ = 1 and num ₂ 2 are set. Then, the num ₁ = 1 person wave attenuation r _{loss, 1} is assigned, num ₂ = each wave attenuation in the equipment 2 of the store r _{loss, 2} is assigned. As a result, the above equation (8) is set as r _{pos, i} + r _{loss, 1} + 2 · r _{loss, 2} .

  Here, the object distribution determination unit 113 is a predetermined position range including an estimated location of a person (an object capable of moving) estimated from object identification information, and in FIG. 4, a circle having a predetermined radius c centered on the estimated location. The range of is the distribution range of this person. Next, when a line segment connecting the beacon 2 and the position pos overlaps the circle range, the person is counted as an object existing between the beacon 2 and the position pos.

  In fact, since a certain degree of error is usually included in the location (image estimated coordinates) of a person or the like estimated by image data of a camera or the like, this estimated position can be determined by appropriately setting the predetermined radius c. It is possible to absorb the error.

  As described above in detail, terminal position determination section 116 calculates presence probability density function p (RSSI | pos) reflecting radio wave attenuation information determined from an object distribution including a movable object such as a person or the like. . Based on the calculated probability density function p (RSSI | pos), the terminal position determination unit 116 performs maximum likelihood estimation, that is, based on the RSSI, the terminal 3 is determined based on which position the terminal 3 is likely to be. Determine the position of

  Thus, according to the terminal position determination unit 116, the existence probability density may be determined in consideration of the presence of an object between the position and the beacon 2 for each position in the position estimation area, It is not necessary to determine the existence of the object between 2 and its number. As a result, even in the stage where the position of the terminal 3 is unknown (undecided), highly accurate terminal position estimation can be performed by calculating an existing probability density function taking in radio wave attenuation information by an object. Furthermore, ID matching technology that links the video flow line ID by performing object recognition from camera image data with the terminal ID of the terminal 3 involved in position estimation becomes unnecessary.

  Returning to FIG. 2, the radio wave source selection unit 116a, as will be described in detail later with reference to FIG. 6, in the form in which a plurality of beacons 2 are arranged, the existence probability density function p (RSSI | Based on pos), a suitable beacon 2 (preferred radio wave source) suitable for position estimation of the terminal 3 is selected from the plurality of beacons 2. The terminal position determination unit 116 uses information (RSSI) related to a plurality of preferred beacons 2 (preferred radio wave source) selected in this way among the acquired received radio wave information (RSSI in the present embodiment), The position of the terminal 3 is determined.

  Furthermore, based on the existence probability density function p (RSSI | pos) for the plurality of selected preferred beacons 2 (preferred radio wave sources), the items pertaining to “pluralitys of the preferred beacons 2 (preferred radio wave sources)” It is also preferable to determine the position of the terminal 3 using received radio wave information (RSSI). Here, the “plurality of suitable radio wave sources” may be selected in order from the highest one of the RSSI of the preferred beacons 2 (preferred radio wave sources) by a predetermined number.

[Terminal position estimation method]
FIG. 5 is a flowchart showing an embodiment of a terminal position estimation process in the object recognition information management unit 112, the object distribution determination unit 113, the radio wave attenuation information determination unit 114, and the terminal position determination unit 116.

(S101) Acquire object arrangement information (e.g., installation position coordinates of facilities in a store, etc.).
(S102) The beacon arrangement information including the arrangement A _set of n beacons 2 arranged in advance and the number m of preferred beacons 2 to be selected is acquired.

  Next, a loop (steps S103 to S111) of suitable beacon selection processing and terminal position determination processing at each time of the position estimation period is entered. That is, the processing of steps S103 to S111 is performed at each time. Thus, the position of the terminal 3 every moment is determined (estimated).

(S103) Object (person) recognition information is acquired.
(S104) Object distribution information is generated from object arrangement information and object (person) recognition information.
(S105) It is determined whether there is a moving or movable object (person). Here, when a false determination (determination that it does not exist) is performed, the process proceeds to step S107 b.

(S107 b) All the n beacons 2 (placement radio wave sources) installed in advance are regarded as suitable beacons 2 (preferred radio wave sources) at this time.
This is particularly effective when the arrangement of the n beacons 2 is preferably determined in advance by the radio wave source arrangement unit 115 based on the object arrangement information. In addition, as another flow form, it is also possible to adopt a form in which the process proceeds to step S107a described later when the determination of false is performed in step S105 after canceling step S107b.

(S106) On the other hand, if a true determination (determination as existing) is performed in step S105, then whether or not the position of a moving or movable object (person) has changed, or It is determined whether the determined object distribution has changed.

  Here, that the position is changed is, for example, a predetermined range including the position at which the position is separated by a predetermined distance or more from the position at the immediately preceding time or including the position at the immediately preceding time. It may be outside. Also, that the object distribution is changing means that the position of at least one object constituting the object distribution is separated from the position at the previous time by a predetermined distance or more, or the previous one. It may be out of a predetermined range including the position at the time of.

If a false determination (determination that there is no change) is performed in step S106, the process proceeds to step S107c.
(S107c) The preferred beacon 2 (preferred radio wave source) selected at the immediately preceding time is set as the preferred beacon 2 (preferred radio wave source) at this time. When there is no suitable beacon 2 selected at the immediately preceding time, the process proceeds to step S107a regardless of the determination result in step S106.

(S107a) On the other hand, when the true determination (determination as changing) is performed in step S106, n pieces of prearranged pieces are allocated using maximum likelihood estimation by the existing probability density function p (RSSI | pos). The m preferred beacons 2 are selected from the beacons 2.
This is processing for excluding the beacon 2 that is more strongly affected by the moving object or the movable object that can be moved (e.g., shielding). The preferred beacon selection process will be described in detail later using FIG.

(S108) Acquire received radio wave information from the terminal 3.
(S109) The predetermined number q (for example, three) are further selected in order from the one with the highest RSSI among the selected preferred beacons 2 (preferred radio wave source).

(S110) A position with the highest existing probability density is obtained by the maximum likelihood estimation method using the existing probability density function p (RSSI | pos) for the RSSI related to the q beacons 2 (the received radio wave strength group is determined in step S108 The position of the terminal 3 (which will be the obtained RSSI) is determined.
In addition, it is also possible to select three beacons 2 (q = 3) higher in the RSSI in step S109 and determine the position of the terminal 3 by three-point positioning using these three. However, in the present embodiment, in order to achieve higher estimation accuracy, terminal position estimation is performed using the same form as the existing probability density function p (RSSI | pos) used in step S107a.
(S111) A terminal position estimation result including the (two-dimensional or three-dimensional) estimated position coordinates in the terminal 3 is output.

  By performing the processing of steps S103 to S111 described above at each time during the position estimation period, it becomes possible for the terminal position determination unit 116 to form time-series information of estimated position coordinates of the terminal 3 . In the present embodiment, the generation of the object distribution information in step S104, that is, the update, and the acquisition of the received radio wave information in step S108, that is, the update are synchronized, and as a result, A fitted preferred maximum likelihood estimation is being performed.

  FIG. 6 is a flowchart showing an embodiment of the preferred radio wave source selection processing in the radio wave source selection unit 116a. The flow shown in the figure corresponds to the process in step S107a of FIG.

In the radio wave source selection unit 116a, first, object distribution information and beacon arrangement information (n beacon arrangements A _set , the number m of suitable beacons to be selected) are set. Further, the arrangement pattern performed one by one the installation of beacons _{2 A i (A i ⊂A set} , A 0 = φ) and the arrangement pattern A _{The set} of the plurality of beacons 2 which are pre-positioned (located radio sources) . That is, the arrangement pattern A _p (p = 1, 2,..., M) is a set of arrangement coordinates of the beacon 2 whose position is determined up to the p-th beacon. Thereafter, the loop of steps S201 to S207 is repeated until the parameter i is incremented by one from zero to m-1.

Also, in the loop of steps S201 to S207, S201 to S205 are executed for each of the positions pos' of all the set coordinates (x ', y') in the position estimation area (ie, a loop is formed ). Furthermore, in the loop of S201 to S205, S201 to S204 are similarly executed for each of all set coordinates (x, y) in the position estimation area (that is, forming a loop). Furthermore, in the loop of S201 to S204, S201 to S203 are executed for each of all the beacons 2 constituting the arrangement pattern A _i (i.e. forming a loop).

(S201) One beacon 2 (related to the current loop) constituting the arrangement pattern A _i (⊂ A _set ) and the position pos of the coordinates (x, y) (related to the current loop) in the position estimation area Calculate the distance r _pos .
(S202) Based on the set object distribution, the radio wave attenuation information num and r _loss between the beacon 2 and the position pos are determined.
(S203) Based on the determined num and r _loss , the existence probability density function p (rssi _pos | pos) with rssi _pos as the RSSI at the position pos by radio waves from the beacon 2 can be expressed by the above equation 4) Calculate using (6).

(S204) Presence of all received radio waves from all the beacons 2 using the existence probability density function p (rssi _pos | pos) calculated for each of all the beacons 2 constituting the arrangement pattern A _i (⊂ A _set ) The probability density function p _Ai (RSSI | pos) is calculated using the above equation (7).

(S205) Then, after the processing of the loop for the coordinates (x, y) position pos including the above steps S201~S204 is completed, the following equation _{(9) E (A i,} pos') = Σ pos∈R | pos'-pos | * p _Ai (RSSI | pos)
The expected value E (A _i , pos ′) of the distance between the positions pos and the position pos ′ (that is, the estimation error of the position) defined by is calculated. In the above equation (9), pos ∈ _R is a summation (summation) of all possible positions pos in the position estimation area.

(S206) Then, after the processing of the loop for the coordinate (x ′, y ′) position pos ′ including the above steps S201 to S205 is completed, the cost function C (A _i ) is further expressed by the following equation (10) C (A _i ) = Σ _pos'∈R E (A _i , pos')
And calculate the cost function C (A _i ) for the placement pattern A _i . In this above equation _{(10), E (A i} , pos ') , all the position pos which can take in position estimation region' is the sum of the (summation). That is, the cost function C (A _i ) defined here is a sum of expected values of position estimation errors in the entire position estimation area.

(S207) to evaluate the arrangement pattern A _i by the calculated cost value of the function C (A _i). Specifically, under the placement pattern A _i , the placement position of the beacon 2 to be placed at the (i + 1) th position is a _{i + 1} , and the following equation (11) a _{i + 1} = arg min C (A _i ∪ { a})
The placement position a _{i + 1} is determined using In the above equation (11), arg min is calculated for all possible arrangement positions a (∈A _set ). Using this arrangement position a _{i + 1} , the arrangement A _{i + 1} is expressed by the following equation (12) A _{i + 1} = A _i ∪ {a _{i + 1} }
Determined by

(S208) As described above, the arrangement position a of the beacon 2 which minimizes the defined cost function C (A _i ) is sequentially determined for all parameters i, and finally, the arrangement positions a _{1 to} a a _m m pieces of suitable beacon placement a _opt (= a _m) is determined. Then, based on the preferred beacon arrangement A _opt , the preferred beacon 2 is selected from among a plurality of beacons 2 placed in advance.

  As described above, in the radio wave source selection unit 116a, one that is more strongly affected by the shielding by an object (person) moving or moving from among the n beacons 2 arranged in advance is It is excluded. By this, after this, the terminal position determination unit 116 can determine the position of the terminal 3 using the suitable beacon 2 more suitable for position determination, and as a result, the conventional method does not consider the influence of the shielding by the moving object etc. As compared with the above, the terminal position estimation accuracy is considerably improved.

  Returning to FIG. 2, the radio wave source placement unit 115 performs the same process as the process in the radio wave source selection unit 116a described above with reference to FIG. Among them, the arrangement position where the beacon 2 is actually arranged in advance is determined. Here, the candidate position in the store may be, for example, a position on a wall surface in the store, a position on equipment such as a desk or a chair, or the like, and may not be set on the floor surface.

Specifically, the radio wave source arranging unit 115 sets _{all the} settable arrangements of n beacons 2 to A _all (⊂S), and
(A) Calculate an existing probability density function p _A (RSSI | pos) for each arrangement pattern A (⊂A _all ) at candidate positions where beacons 2 can be installed in the store,
(B) Generate a cost function C (A) relating to the distance between the position pos and the position pos',
(C) Determine the arrangement A _set (⊂A _all ) of n beacons 2 that minimizes the generated cost function C (A).

Thereby, the arrangement of the appropriate beacons 2 in consideration of the arrangement of the objects installed in advance, which is the premise of the preferable beacon selection process in the radio wave source selection unit 116a, is determined. Note that although the process in the radio wave source placement unit 115 described above is the same as the process in the radio wave source selection unit 116a, the set of beacon placement candidate points input for the process is not A _opt but A _all Differs in that

  Similarly, in FIG. 2, the terminal flow line information management unit 117 collects estimated position coordinates of the terminal 3 for each time during the position estimation period, which is determined by the terminal position determination unit 116, and the terminal 3 (user who possesses) For each of the identifier IDs, flow line information is generated in which time series information of the collected estimated position coordinates is associated. In addition, the terminal flow line information management unit 117 may store the generated flow line information in the terminal flow line information storage unit 105, and it is also preferable to display the generated flow line information on the display 106 via the input / output control unit 123. It may be transmitted to an external information processing apparatus via the control unit 121 and the communication interface 101.

  As described above in detail, according to the present invention, the terminal position is determined using the existing probability density function having a variable reflecting radio wave attenuation information. Here, the radio wave attenuation information is determined based on an object distribution that may include a movable object such as a person. As a result, even in the situation where movable objects may exist, it is possible to consider how much the received radio wave is attenuated by the “object distribution that may include movable objects”, and the radio waves from the radio wave source The position of the received terminal can be estimated more accurately.

  Further, according to the present invention, even in the embodiment in which the object recognition means assigns the video flow line ID to a movable object such as a person and tracks the movement, the position estimation target terminal is of any video flow line ID. The position of this terminal can be determined using an existing probability density function reflecting the object distribution without considering whether it is linked to an object. That is, it is possible to carry out terminal position estimation processing more simply without the need for matching technology between terminal ID and video flow line ID, and at the same time, there is a problem that reliability of presence probability density in maximum likelihood estimation is lowered. It also comes to solve.

  Further, for example, the object distribution is determined based on the object recognition information and the object arrangement information at each time of the position estimation period, and the position of the terminal is estimated every moment by selecting a suitable radio wave source for each time. You can also. Therefore, for example, even in the situation where many persons are moving around, the position of each terminal can be determined with higher accuracy.

  Incidentally, the configuration and method of the present invention are, for example, a monitoring system that monitors places where a large number of people move, stay or move in and out, and further, in one store, in a commercial service facility, or even in a store. The present invention is applicable to various systems such as entering and leaving a person on a street of a city, taking a break, watching a game, participating in an event, and a marketing research system for investigating the movement situation.

  In the various embodiments of the present invention described above, various changes, modifications and omissions of the scope of the technical idea and the aspect of the present invention can be easily made by those skilled in the art. The above description is merely an example and is not intended to be limiting in any way. The present invention is limited only as defined in the following claims and the equivalents thereto.

DESCRIPTION OF SYMBOLS 1 Terminal position estimation apparatus 101 Communication interface 102 Received radio wave information storage part 103 Object recognition information storage part 104 Object arrangement information storage part 105 Terminal flow line information storage part 106 Display ・ keyboard (DP, KB)
111 object recognition unit 112 object recognition information management unit 113 object distribution determination unit 114 radio wave attenuation information determination unit 115 radio wave source arrangement unit 116 terminal position determination unit 116 a radio wave source selection unit 117 terminal flow line information management unit 121 communication control unit 122 received radio wave Information management unit 123 Input / output control unit 2 Beacon (radio wave source)
3 terminal 4 camera

Claims (13)

A terminal position estimation apparatus for acquiring received radio wave information at a terminal that has received radio waves from a radio wave source and estimating the position of the terminal,
The object recognition information from the object recognition means capable of recognizing the movable object is acquired, and the movable object and the previously installed object are obtained based on the object recognition information and the arrangement information of the previously installed object Object distribution determining means for determining an object distribution which is a distribution of objects;
Radio wave attenuation information determination means for determining radio wave attenuation information related to the degree of radio wave attenuation by the object between the radio wave source and the position in the position estimation region based on the object distribution determined in the position estimation region ,
The position of the terminal is determined based on the existence probability density function which is a presence probability density function for the position in the position estimation area and the amount of the received radio wave reflecting the determined radio wave attenuation information as a variable And a terminal position determination unit.   The amount related to the received radio wave reflecting the radio wave attenuation information is an amount indicating the number of the objects existing between the radio wave source and the position and the degree of radio wave attenuation by each object, and the reference received radio wave intensity to be a reference The terminal position estimation apparatus according to claim 1, wherein the terminal position estimation apparatus is a received radio wave intensity that reflects radio wave attenuation by the object, which is calculated from   The terminal position determination means adopts at least a corresponding predetermined amount for each of the movable object and the previously installed object as an amount indicating the degree of the radio wave attenuation. The terminal position estimation device according to.   The object distribution determination means sets a predetermined position range including the estimated location of the movable object estimated from the object identification information as the distribution range of the movable object. The terminal position estimation device according to any one of 1 to 3. The radio wave source is a plurality of radio wave sources arranged in advance.
The terminal position determination means selects a suitable radio wave source suitable for position estimation of the terminal from among the plurality of radio wave sources based on the existence probability density function, and is selected from among the received radio wave information acquired. The terminal position estimation apparatus according to any one of claims 1 to 4, wherein the position of the terminal is determined using information on a plurality of the suitable radio wave sources.   The terminal position determination means uses the received radio wave information related to a plurality of the suitable radio wave sources based on the existence probability density function for a plurality of the selected suitable radio wave sources, and the position of the terminal The terminal position estimation apparatus according to claim 5, characterized in that   The terminal position estimation apparatus according to claim 6, wherein a plurality of the suitable radio wave sources are further selected by a predetermined number in order from the highest radio wave intensity among the suitable radio wave sources. .   The terminal position determination means selects the preferred radio wave source selected at the previous time when the object distribution at one time related to position estimation has not changed by a predetermined amount or more as compared with the object distribution at the previous time. The terminal position estimation apparatus according to any one of claims 5 to 7, characterized in that is a suitable radio wave source at the one time.   The terminal position determination means is characterized in that, when there is no movable object at one time for position estimation, all of the plurality of arranged radio wave sources are used as suitable radio wave sources at the one time. Item 9. The terminal position estimation device according to any one of Items 5 to 8.   The object recognition information is generated based on imaging or measurement information from at least one of a camera, an omnidirectional camera, a depth camera, an infrared camera, an infrared positioning unit, a laser positioning unit, and a thermographic unit. The terminal position estimation apparatus according to any one of claims 1 to 9, wherein   At least for the object arrangement related to the arrangement information of the objects installed in advance, the arrangement of the radio wave source that minimizes the error of the estimated position of the terminal is determined based on the existence probability density function, and the arrangement is determined based on the determined arrangement. The radio wave source arrangement means according to any one of claims 1 to 10, further comprising radio wave source arrangement means for determining an arrangement position in which a plurality of radio wave sources are arranged beforehand from among candidate positions where radio wave sources can be arranged. Terminal position estimation device. A program that causes a computer mounted on a device to obtain received radio wave information at a terminal that has received radio waves from a radio wave source and estimate the position of the terminal,
The object recognition information from the object recognition means capable of recognizing the movable object is acquired, and the movable object and the previously installed object are obtained based on the object recognition information and the arrangement information of the previously installed object Object distribution determining means for determining an object distribution which is a distribution of objects;
Radio wave attenuation information determination means for determining radio wave attenuation information related to the degree of radio wave attenuation by the object between the radio wave source and the position in the position estimation region based on the object distribution determined in the position estimation region ,
The position of the terminal is determined based on the existence probability density function which is a presence probability density function for the position in the position estimation area and the amount of the received radio wave reflecting the determined radio wave attenuation information as a variable A terminal position estimation program characterized by causing a computer to function as terminal position determination means. A terminal position estimation method using a computer mounted on an apparatus for acquiring received radio wave information at a terminal that has received radio waves from a radio wave source and estimating the position of the terminal,
The object recognition information from the object recognition means capable of recognizing the movable object is acquired, and the movable object and the previously installed object are obtained based on the object recognition information and the arrangement information of the previously installed object Determining an object distribution, which is a distribution of objects;
Determining radio wave attenuation information according to the degree of radio wave attenuation by the object between the radio wave source and the position in the position estimation region based on the object distribution determined in the position estimation region;
The position probability of the terminal is determined based on the presence probability density function with the amount of the received radio wave information reflecting the determined radio wave attenuation information as a variable. A terminal position estimation method comprising the steps of:

Images