WO2011013245A1 - Position estimating device - Google Patents

Abstract

A position estimating device is provided with a positioning section (111) which positions a current position; a detecting section (112) which detects the current action state of a user; a first scene estimating section (113) which estimates the current life scene of the user by means of a life scene corresponding to the current position, the current action state, and a current time from first estimation information containing life scenes being action scenes in the life of the user, time zones, positions, and the action states of the user; a second scene estimating section (115) which estimates a future life scene of the user by means of the next scene corresponding to the current life scene from second estimation information containing the life scenes and next scenes; and a first position estimation section (116) which estimates the kind of a facility where the user will be present a predetermined time later by means of the kind of a facility corresponding to the future life scene from first association information associating the life scenes with the kinds of facilities where the user is present in the life scenes.

Description

Position estimation device

The present invention relates to estimation of a position where a user is located.

2. Description of the Related Art Conventionally, a technique for estimating a position where a user is located after a predetermined time using a current position of a user who possesses a position estimation device is known. For example, a destination prediction apparatus that predicts a destination by reflecting user behavior characteristics is disclosed (see Patent Document 1). In the movement destination prediction apparatus of Patent Literature 1, the position information and date / time of the user are detected and the movement history of the user is accumulated. Then, the frequency of arrival at the destination (arrival frequency) and the frequency of arrival at each day of the week within the predetermined period (appearance frequency) are calculated from the accumulated movement history, and based on the calculated arrival frequency and appearance frequency The action class is calculated by combining one day of the week whose arrival frequency is equal to or higher than the threshold and which is higher than the threshold. Then, based on the accumulated movement history, the calculated action class, and the current position information of the user, a destination to which the user is going from now is predicted.

JP 2009-36594 A

However, in the destination prediction apparatus of Patent Document 1, since the destination is predicted using the movement history and action class accumulated in advance, the location where the user is located after a predetermined time is estimated at the place visited for the first time. I couldn't.

The present invention has been made in view of the above, and by using a life scene that is an operation scene in a user's life, the position where the user is located after a predetermined time can be estimated even in a place visited for the first time. An object is to obtain a position estimation device.

In order to solve the above-described problems and achieve the object, the present invention provides a life scene that is an operation scene in a user's life, a time zone of the life scene, a position of the life scene, and the life scene. The first estimation storage unit that stores first estimation information having the operation state of the user in the first storage information, the first correspondence information in which the life scene is associated with the facility type where the user is located in the life scene is stored A second correspondence storage unit that stores second estimation information including a first correspondence storage unit, a life scene, and a next scene representing a scene of an operation performed by the user next to the life scene, and a current position From the positioning unit that performs positioning, the detection unit that detects the current operation state of the user, and the first estimation information, using the life scene corresponding to the current position, the current operation state, and the current time, Profit A first scene estimation unit that estimates a user's current life scene, and the second scene, and the next scene corresponding to the current life scene, and the user's current life scene after the predetermined time. A first scene estimation unit that estimates a future life scene, and a first type that estimates a facility type where a user is located after a predetermined time from the first correspondence information, using the facility type corresponding to the future life scene. A position estimation unit.

According to the present invention, it is possible to estimate the position where the user is located after a predetermined time even in a place visited for the first time by using a life scene that is an action scene in the life of the user.

The external appearance block diagram which shows an example of a portable terminal. The block diagram which shows an example of a functional structure of a portable terminal. The figure which shows an example of 1st estimation information. The figure which shows an example of 2nd corresponding | compatible information. The figure which shows an example of 1st corresponding | compatible information. The figure which shows an example of 2nd estimation information. The figure which shows an example of the display of an estimation result. The figure which shows an example of the display of an estimation result. The flowchart which shows the flow of the position estimation process by a portable terminal. The flowchart which shows the flow of the position estimation process by a portable terminal. The block diagram which shows an example of a functional structure of a portable terminal. The figure which shows an example of 1st corresponding | compatible information. The flowchart which shows the flow of the position estimation process by a portable terminal. The block diagram which shows an example of a functional structure of a portable terminal. The figure which shows an example of a movement history. The figure which shows an example of a movement pattern. The flowchart which shows the flow of the position estimation process by a portable terminal.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a preferred embodiment of a position estimation device according to the present invention will be described in detail with reference to the accompanying drawings. In the following embodiment, an example in which the position estimation device of the present invention is applied to a mobile terminal such as a mobile phone or a PDA (Personal Digital Assistant) is shown, but the present invention is not limited to this and is portable. Any device that can be equipped with a positioning function and a function for detecting the operating state of the owner is applicable.

(Embodiment 1)
FIG. 1 is an external configuration diagram illustrating an example of a mobile terminal according to the first embodiment. At least one positioning sensor and at least one body motion sensor are connected to one mobile terminal 100.

The positioning sensor 101 is a sensor that receives radio waves from GPS satellites at predetermined time intervals in order to measure the current position of the mobile terminal 100. In the present embodiment, the positioning sensor 101 uses a GPS function as a typical configuration. However, the positioning is not limited to the positioning using the GPS function. For example, when the mobile terminal 100 is a mobile phone, the positioning sensor 101 is configured as an antenna that receives radio waves from a communication base station, and the received radio waves are emitted. You may comprise so that it may measure by specifying the communication base station which performed. Further, the positioning sensor 101 may be configured as an antenna that receives radio waves from a wireless LAN (Local Area Network) access point, and may be configured to perform positioning by specifying the access point that has emitted the received radio waves. . Further, the positioning result does not necessarily need to be acquired by the mobile terminal 100, and the user may have an RFID (Radio Frequency IDentification) tag, and positioning may be performed by an RFID antenna that is an external system and an associated system. .

The body motion sensor 102 is a sensor that detects a signal corresponding to the operation of the user holding the mobile terminal 100 at a predetermined time interval. In this embodiment, the body motion sensor 102 has a three-axis acceleration as a typical configuration. It is comprised as an acceleration sensor which detects the acceleration data according to a user's operation | movement using. However, the present invention is not limited to this, and the body motion sensor 102 may be constituted by, for example, a gyroscope, a direction sensor, or the like.

In the present embodiment, a case where one user uses the mobile terminal 100 will be described, but the present invention is not limited to this, and a single mobile terminal may be used by a plurality of users.

FIG. 2 is a block diagram of an example of a functional configuration of the mobile terminal according to the first embodiment. As illustrated in FIG. 2, the mobile terminal 100 includes a positioning unit 111, a detection unit 112, a first scene estimation unit 113, a storage unit 114, a second scene estimation unit 115, and a first position estimation unit 116. The scene update unit 117, the output unit 118, the first estimation storage unit 131, the first correspondence storage unit 142, the second estimation storage unit 132, and the second correspondence storage unit 141 are mainly provided. The mobile terminal 100 according to the first embodiment further has a hardware configuration such as a CPU, a storage medium such as an HDD or a memory (not shown).

The positioning unit 111 measures the current position of the mobile terminal 100 using the radio wave received from the GPS satellite by the positioning sensor 101.

The detection unit 112 acquires current acceleration data from the body motion sensor 102, and detects the current operation state (body movement) of the user who holds the mobile terminal 100 from the acquired acceleration data. Specifically, the detection unit 112 first estimates the direction of gravity from the acquired acceleration data. The estimation of the direction of gravity is performed by calculating the average of acceleration data for a certain time as a three-dimensional vector and then normalizing the vector so as to have a magnitude of 1G. Thereafter, the detection unit 112 calculates acceleration data excluding the gravity component by subtracting the gravity vector from the acceleration data. Further, the detection unit 112 calculates three feature amounts including the length of the obtained acceleration data, the inner product of the acceleration data and the gravity vector, and the outer product. These feature values are numerical values that are not affected by the orientation of the mobile terminal 100 equipped with the acceleration sensor, regardless of the direction in which the user holds the mobile terminal 100. It can be used as a feature amount for obtaining the context of the moving means.

Next, the detection unit 112 calculates a statistical feature value from these three feature values. The statistical feature amount is the maximum value, minimum value, average, or variance of the feature amount during a certain period of time. By calculating these four values for each feature quantity, a 12-dimensional value can be acquired. This statistical feature quantity can be used as a feature quantity that takes into account the temporal transition of acceleration. Next, the detecting unit 112 applies a neural network to the obtained 12-dimensional value, so that the current user is in any one of four states of running / walking / riding the vehicle / stationary Guess that. It is assumed that the neural network has been learned by giving statistical data in each of these four states acquired in advance as correct data. Each of the four states is represented by a value from 0 to 1, and the larger the numerical value, the higher the degree that the state is estimated. As described above, the detection unit 112 can detect four states of “running”, “walking”, “riding on a vehicle”, and “still” based on the value of acceleration data from the acceleration sensor (body motion sensor 102). .

In the present embodiment, the user's operation state is detected based on the acceleration data detected by the body motion sensor 102. However, the present invention is not limited to this. The operation state of the user may be detected by estimating the operation state of the user holding the mobile terminal 100 with a certain accuracy based on the signal or the image pickup signal from the camera.

The first estimation storage unit 131 is a storage medium such as an HDD (Hard Disk Drive) or a memory that stores first estimation information for estimating the current life scene of the user. Here, the life scene is, for example, a scene of an operation performed by the user in the user's life such as “work”, “business trip”, “dinner”, “shopping”, and “on the way home”, that is, an action in the user's life. Indicates the situation. Further, the first estimation information is registered in advance in the mobile terminal 100.

FIG. 3 is a diagram illustrating an example of first estimation information for estimating a user's life scene. The first estimation information shown in FIG. 3 is obtained by registering an individual user based on the position of the user's home, work place, frequently visited store, etc., and the time zone where the user is located. This is a generated table. Here, the first estimation information is stored in a table format, but the present invention is not limited to this and may be stored in another format.

As illustrated in FIG. 3, the first estimation information includes a life scene that is an operation scene in a user's life, a time zone that is the life scene, a position of the life scene, and a user's life scene. The operation states are associated with each other, and 1 to 12 are stored in descending order of priority. For example, if the life scene is “working”, the user can select “stationary” or “latitude” at the position of “35.495500, 139.593122” in the time zone “8:00 to 22:00”. You can see that you are “walking”. Here, in the first estimation information of FIG. 3, since the position indicates the central latitude and longitude, the distance from the actual user position is calculated, and if it is within a certain range, for example, within 100 m. It is assumed that the user is located.

The second correspondence storage unit 141 is a storage medium such as an HDD or a memory that stores second correspondence information for estimating the position where the user is located. FIG. 4 is a diagram illustrating an example of second correspondence information in which a facility type is associated with a position of the facility type. As shown in FIG. 4, the second correspondence information is a table in which the facility type indicating the type of facility, the position of the facility type, and the probability that the user is located at the position of the facility type are associated with each other. . For example, it can be seen that the probability that the user is located in the “company” of the facility type at the position where the latitude and longitude are “35.495500, 139.593122” is “17/20”. Here, the second correspondence information is stored in a table format, but the present invention is not limited to this and may be stored in another format.

The first correspondence storage unit 142 is a storage medium such as an HDD or a memory that stores first correspondence information for estimating the facility type where the user is located. FIG. 5 is a diagram illustrating an example of first correspondence information in which a life scene is associated with a facility type. As shown in FIG. 5, the first correspondence information is a table in which a life scene, a facility type where the user is located in the life scene, and a probability that the user is located in the facility type are associated with each other. For example, when the life scene is “dinner”, the probability that the user is located in the “type restaurant” of the facility type is “12/20”. Here, the first correspondence information is stored in a table format, but the present invention is not limited to this and may be stored in another format.

In addition, the first correspondence information described above is preferably constructed and updated from the user's personal history, but this is not a limitation. By using general data prepared by the creator of the mobile terminal 100 in advance, a future position with a certain accuracy may be estimated from the beginning.

The second estimation storage unit 132 is a storage medium such as an HDD or a memory that stores second estimation information for estimating a future life scene that is a life scene after a predetermined time of the user. FIG. 6 is a diagram illustrating an example of second estimation information in which the current life scene is associated with the next scene that is the next life scene. As shown in FIG. 6, the second estimation information includes a life scene, a shortest time in which the life scene is continued, an average time, a longest time, and a scene of an operation performed by the user next to the life scene. Is a table having a next scene representing Further, the probability that the user's action is actually the next scene is attached to the next scene. For example, when the current life scene is “working”, the shortest time for which work has been continued is “4 hours”, the average time is “9 hours”, and the longest time is “14 hours”, and the next life scene is It can be seen that the probability of “dinner” is 6/16, the probability of “returning on board” is 4/16, and the probability of “shopping” is 2/16. Here, the second estimation information is stored in a table format, but the present invention is not limited to this and may be stored in another format. Further, the second estimation information is registered in advance in the mobile terminal 100.

The first scene estimation unit 113 estimates the current life scene of the user who owns the mobile terminal 100. Specifically, the first scene estimation unit 113 uses the first estimation information to determine the current position immediately followed by the positioning unit 111, the current operation state detected by the detection unit 112, and the time measurement unit (not shown). Of the life scenes corresponding to the current time, a life scene with a high priority is estimated as the user's current life scene.

Note that the first scene estimation unit 113 uses the current operation state to estimate the life scene, and thus the operation state such as walking or stationary that cannot be determined by the positioning function is clarified. Even in situations where positioning is difficult, it can be estimated that the user is on board. Also, the first scene estimation unit 113 becomes “estimation impossible” when it does not correspond to any life scene. Here, the current life scene is estimated from the current position measured by the positioning unit 111 and the operation state detected by the detection unit 112 using the first estimation information. However, the present invention is not limited to this. It is good also as a structure estimated from the electronic data of a schedule, operation of a portable terminal, and the log | history of communication.

The storage unit 114 acquires the facility type corresponding to the current position measured by the positioning unit 111 from the second correspondence information, and obtains the current life scene estimated by the first scene estimation unit 113 and the acquired facility type. The first correspondence information in the first correspondence storage unit 142 is updated by storing the first correspondence information in association with each other. Further, the storage unit 114 updates the probability of being located in the facility type in the first correspondence information and the probability of being located in the position of the facility type in the second correspondence information based on the transition of the life scene of the user.

The second scene estimation unit 115 uses the next scene corresponding to the current life scene estimated by the first scene estimation unit 113 based on the second estimation information, after a predetermined time of the user who owns the mobile terminal 100. A future life scene that is a life scene is estimated.

Specifically, first, the second scene estimation unit 115 estimates the end time of the current life scene by storing the current life scene and the time when the life scene started. Here, the average time in FIG. 6 is applied as the life scene time. For example, if the current life scene is “working” and 8 hours and 40 minutes have already passed since becoming the life scene, referring to the average time, the life scene of “working” continues for 9 hours. It can be estimated that it will be completed in about 20 minutes.

Here, when the mobile terminal 100 estimates a future life scene after 30 minutes, the probability that the mobile terminal 100 has transitioned to the “dinner” life scene is 6/16, and the life scene is “on the way home”. 4/16, and the probability of transition to the “shopping” life scene is 2/16. Although not shown in FIG. 6, the same applies when there are four or more next scenes. Further, the probability can be corrected more accurately by correcting the probability using the shortest time and the longest time, or by holding the end time, the distribution thereof, and the like in the second estimation information.

The first position estimation unit 116 estimates the facility type corresponding to the future life scene from the first correspondence information as the facility type where the user holding the mobile terminal 100 is located after a predetermined time. Then, the first position estimation unit 116 multiplies the probability that the second estimation information is the next scene corresponding to the current life scene and the probability that the first correspondence information is located in the facility type corresponding to the future life scene. Is calculated as the probability that the user is located in the estimated facility type.

For example, in the above-mentioned example, the probability that the user's future life scene after 30 minutes has changed to the life scene of “dinner” is 6/16, and the user changes to the life scene of “on the way home”. 4/16, and the probability of transition to the “shopping” life scene is 2/16. Then, using the first correspondence information in FIG. 5, the first position estimation unit 116 calculates the probability P _type located in each facility type corresponding to those future life scenes after a predetermined time by the following equation. .

　Ｐ_type：所定時間後に利用者が施設種別typeに所在する確率
　Ｐ_scene：所定時間後の将来生活シーンscene発生確率
　Ｐ(scene,type)：将来生活シーンscene時に対応する施設種別typeが利用される確率

P _type : Probability that the user is located in the facility type type after a predetermined time P _scene : Probability of occurrence of a future life scene after a predetermined time P (scene, type): The facility type type corresponding to the future life scene is used probability

In the example as described above, the first position estimation unit 116 indicates that the probability that the user who owns the mobile terminal 100 is located in the facility type “railway” with the life scene “on the way home” after 30 minutes is 1 /. 4. Life scene is “Dinner” and the probability of being in the “Restaurant” facility type is 9/40, Life scene is “Dinner” and the probability of being in the “Company” facility type is 3/20, and the life scene is “ The probability of being in the facility type “shop” in “shopping” is calculated as 1/8.

In addition, the first location estimation unit 116 further determines the location of the facility type corresponding to the facility type located after the calculated predetermined time from the second correspondence information, and the user holding the mobile terminal 100 is located after the predetermined time. Estimated as the location of the facility type. Then, the first position estimation unit 116 has the probability that the second estimation information is the next scene corresponding to the current life scene, the probability that the first correspondence information is located in the facility type corresponding to the future life scene, 2 The product of the probability that the user is located at the location of the facility type corresponding to the facility type located after a predetermined time in the correspondence information is calculated as the probability that the user is located at the estimated location of the facility type. In other words, the first position estimating unit 116 can calculate the specific position from the estimated facility type by using the second correspondence information shown in FIG. 4 and calculate the probability that the user is located at the position. .

For example, in the second correspondence information in FIG. 4, the probability of using the restaurant at the position “35.551331, 139.6675131” is 6/8, and the restaurant at the position “35.552345, 139.6671251” is used. Therefore, the probability of the location of the facility type is calculated by multiplying these probabilities by the probability of being located in the facility type. That is, in the above-described example, the probability that the user is located in the former restaurant after 30 minutes is 27/160 by multiplying the probability 9/40 located in the facility type of “restaurant” by the probability 6/8. Similarly, the probability of being in a restaurant can be calculated as 9/160 by multiplying the probability 9/40 by the probability 2/8. The first position estimation unit 116 may be configured to correct the probability calculated according to the current position.

The scene update unit 117 updates the first estimated information registered in advance according to the position measured when the user actually lives, the detected operating state, and the time zone counted.

For example, when updating the first estimation information in FIG. 3, the scene update unit 117 has a time zone corresponding to the life scene “work” from 8:00 to 22:00 from the first estimation information. It is assumed that the user's operation state remains “still” and continues until 23:00 even after “work” is continued until that time. Thus, it can be presumed that the user's operation state is continued “still” and that the previous life scene is continued. That is, in this case, the user presumes that “work” has continued until 23:00, and updates the time zone corresponding to “work” in the first estimation information to 8-23 (8 to 23:00). .

In addition, the scene update unit 117 updates the probability that the user is located in the facility type in the second estimated information registered in advance with the life scene when the user actually lives.

For example, when updating the second estimation information in FIG. 6, the scene update unit 117 determines that the life scene changes to “on boarding” after the life scene is estimated to be “working” at a certain time. The probability of the next scene corresponding to the “work” life scene in the second estimation information is updated. In other words, the denominator of the probability of all the next scenes corresponding to “work” is increased by 1 and updated to 17, and the numerator is also increased by 1 for the next scene of “on the way home” that has actually shifted from 4/16. Update to 5/17.

The output unit 118 uses the facility type estimated by the first position estimating unit 116 and the certainty factor calculated by the first position estimating unit 116 as the probability that a user exists in the facility type as an estimation result. Output to a display unit (not shown). FIG. 7 is a diagram illustrating an example of an estimation result display. As illustrated in FIG. 7, the certainty factor that is the probability of existing in the facility type is displayed in association with each facility type estimated by the first position estimating unit 116. For example, when the probability that the user is located on the railway is estimated to be ¼, the probability (certainty factor) “¼” located on the “railway” is displayed in association with the facility type “railway”.

The output unit 118 further has a certainty factor that is the probability that a user exists at the location of the facility type calculated by the first location estimation unit 116 and the location of the facility type estimated by the first location estimation unit 116. Are output to a display unit (not shown) as an estimation result. FIG. 8 is a diagram illustrating an example of the display of the estimation result. As illustrated in FIG. 8, the name of the facility type and the name of the facility type are associated with each facility type estimated by the first position estimation unit 116. The certainty level, which is the probability of being present at the location or location of the facility type, is displayed. For example, when the probability that the user is located in a restaurant at the position “35.551331, 139.675131” is estimated to be 27/160, the restaurant name “Restaurant” is associated with the facility type “Restaurant”. A ”,“ 35.551331, 139.675131 ”which is the location of the restaurant, and the probability (certainty factor)“ 27/160 ”of being located in the restaurant.

Next, position estimation processing by the mobile terminal 100 according to the present embodiment configured as described above will be described. FIG. 9 is a flowchart of a position estimation process performed by the mobile terminal according to the first embodiment. Below, the process in the case of calculating the probability that a user is located in the facility type estimated after the predetermined time will be described.

First, the positioning unit 111 measures the current position of the mobile terminal 100 (step S11). And the detection part 112 detects the present operation state of the user who has the portable terminal 100 (step S12).

Next, the first scene estimation unit 113 estimates the current life scene based on the current position, the current operation state, and the current time based on the first estimation information (step S13). The storage unit 114 updates the first correspondence information with the facility type corresponding to the current position in the second correspondence information and the current life scene (step S14). Then, the second scene estimation unit 115 estimates the future life scene after a predetermined time of the user based on the current life scene based on the second estimation information (step S15).

Next, the first position estimating unit 116 estimates the facility type where the user is located after a predetermined time based on the future life scene based on the first correspondence information, and calculates the probability that the user is located in the facility type. (Step S16). The output unit 118 outputs the estimated facility type and the calculated probability to the display unit as an estimation result (step S17).

Next, the processing when the probability that the user is located at the location of the estimated facility type after a predetermined time is calculated will be described below. FIG. 10 is a flowchart of a position estimation process performed by the mobile terminal according to the first embodiment. In addition, the position estimation process in FIG. 9 or the position estimation process in FIG. 10 receives and accepts a setting input for performing any position estimation process by the user from an operation unit (not shown) of the mobile terminal 100, for example. One of the position estimation processes is performed based on the setting.

First, the process from the current position measurement to the estimation of the future life scene (steps S31 to S35) is the same as the process (steps S11 to S15) in FIG.

Next, the first position estimating unit 116 estimates the facility type where the user is located after a predetermined time based on the future life scene based on the first correspondence information, and calculates the probability that the user is located in the facility type. (Step S36). Then, the first position estimation unit 116 estimates the position of the facility type based on the second correspondence information, and calculates the probability that the user is located at the position of the facility type (step S37). The output unit 118 outputs the estimated location of the facility type and the calculated probability to the display unit as an estimation result (step S38).

As described above, in the mobile terminal 100 according to the first embodiment, the current life scene of the user is estimated from the current position, the current operation state of the user, and the current time, and the estimated current life scene is obtained. Based on this, a future life scene after a predetermined time is estimated. Then, the mobile terminal 100 estimates the location of the facility type or the facility type where the user is located after a predetermined time based on the estimated future life scene, and the user is located at the estimated location of the facility type or the facility type. Probabilities are calculated and output to the display unit. In this way, by using the current life scene of the user, the facility type or the user is likely to be located after a predetermined time from the continuity of the life scene without estimating from the continuity of the position of the user. Since the location of the facility type can be estimated, the facility type of the user or the location of the facility type after a predetermined time can be estimated even at a place visited for the first time.

That is, even when the user's movement history for the place visited for the first time is not stored, the position after a predetermined time can be estimated in consideration of the user's behavior pattern. This is because there is a high possibility that it is common in a place different from the behavior in the life of a certain person. For example, it is presumed that the place visited for the first time will become a dinner scene after a business trip scene, and this user uses a canteen restaurant with a probability of 50% and a ramen restaurant with a probability of 30%. Based on this probability, the position where the user is located after a predetermined time can be estimated.

The life scene of the above embodiment is not limited to the above example. For example, the life scene of “dinner” can be further subdivided into “dinner”, “meal dinner”, “banquet” and the like. In that case, for example, by detecting a companion in the detection unit 112, it is estimated that there is a high possibility that a “banquet” life scene will occur after a predetermined time, and based on this, the first position estimation unit 116 performs a predetermined time It can be estimated that there is a high possibility of being located in “Izakaya” or “Yakitori-ya” later. In other words, it is possible to estimate the future position in consideration of the action state such as “the presence or absence of a companion”.

In the above embodiment, the estimation of the future life scene has been described above. However, in order to estimate the future life scene with higher accuracy, not only the current life scene but also the current action state, It is effective to estimate including the position, current time, day of the week, and the like. For this purpose, a method of retaining the current action state, current position, facility type currently located, current time, day of the week, etc. in the second estimation information is effective. Further, regarding the estimation of the future life scene, other techniques such as a Bayesian network and a hidden Markov model may be used. In this case, as described above, not only the current life scene but also the current behavioral state, the current position, the type of facility currently located, the current time, the day of the week, etc. A future life scene after a predetermined time is estimated using behavior information, state information, and environment information.

Furthermore, when estimating the location of the facility type where the user is located after a predetermined time, it may be configured to use the current location that has been measured and perform estimation considering the reachability to the location. Specifically, for example, when the current life scene is estimated to be “working” and the future life scene after 30 minutes is estimated to be “dinner” with a probability of 6/16, in the above embodiment, The position where the user takes dinner with high probability was calculated and output as an estimation result. On the other hand, when the relationship between the current position and the estimated position after a predetermined time is difficult to move in 30 minutes, it is appropriate to perform a different estimation. That is, the possibility of movement within a specified time (in this example, 30 minutes) is calculated from the positional relationship between the latitude and longitude of the current position and the estimated position after a predetermined time, and unreachable positions are removed. .

For example, when the current position is “35.495500, 139.593122”, the estimated travel time to the restaurant “35.551331, 139.675131” is 50 minutes and cannot be reached within 30 minutes. In this case, the candidate is removed, the probability is similarly calculated using other positions, and the estimation result is output. Note that the estimated travel time at this time may be calculated using a past travel time, or may be calculated using an existing calculation method such as a navigation service.

(Embodiment 2)
In the first embodiment, the facility type where the user is located after a predetermined time is estimated by the first correspondence information in which the life scene is associated with the facility type. On the other hand, in this embodiment, the facility type where the user is located after a predetermined time is estimated based on the first correspondence information in which the life scene and the plurality of facility types are hierarchically associated.

FIG. 11 is a block diagram of an example of a functional configuration of the mobile terminal according to the second embodiment. As illustrated in FIG. 11, the mobile terminal 200 includes a positioning unit 111, a detection unit 112, a first scene estimation unit 113, a storage unit 114, a second scene estimation unit 115, and a first position estimation unit 216. The scene update unit 117, the output unit 118, the first estimation storage unit 131, the second estimation storage unit 132, the second correspondence storage unit 141, and the first correspondence storage unit 242 are mainly provided. Note that the portable terminal 200 according to the second embodiment has a hardware configuration such as a CPU, a storage medium such as an HDD or a memory (not shown), as in the first embodiment. Here, since the same reference numerals as those in the first embodiment are the same as those in the first embodiment, the description thereof is omitted. The external configuration of the mobile terminal 200 according to the second embodiment is the same as that of the first embodiment.

The first correspondence storage unit 242 is a storage medium such as an HDD or a memory that stores first correspondence information for estimating the type of facility where the user is located. FIG. 12 is a diagram illustrating an example of first correspondence information in which a life scene is associated with a facility type. As shown in FIG. 12, the first correspondence information includes the life scene, the name of the facility in the facility type where the user is located in the life scene, and the probability that the user is located in the facility type indicated by the name of the facility. And the category of the facility in the facility type where the user is located in the life scene and the probability that the user is located in the facility type indicated by the category of the facility. The name of the facility is a facility type that is included in the facility category and further classifies the facility category. For example, when the life scene is “business trip”, the probability that the user is located in the facility type “second factory” indicated by the name of the facility is “3/4”, and the facility type indicated by the name of the facility “third” It can be seen that the probability that the user is located in the “factory” is “1/4” and the probability that the user is located in the facility type “factory” indicated by the facility category is “4/4”.

In addition, although the name of the facility and the category of the facility are described here in two layers, it can be realized as three or more layers. In addition, by assigning a label called a facility type tag to each facility, it is also possible to use a label group that is not a complete tree structure and whose hierarchical level is not constant. Here, the first correspondence information is stored in a table format, but the present invention is not limited to this and may be stored in another format.

Similar to the first embodiment, the first position estimation unit 216 estimates the facility type corresponding to the future life scene from the first correspondence information as the facility type where the user who owns the mobile terminal 200 is located after a predetermined time. Then, the probability that the user exists in the estimated facility type is calculated. Since this calculation method is the same as that of the first embodiment, description thereof is omitted.

In addition, when estimating the facility type where the user is located after a predetermined time, the first position estimating unit 216 first estimates the name of the facility that classifies the facility type in detail. Then, the first position estimation unit 216 is located in the probability of the next scene corresponding to the current life scene in the second estimation information and the facility type indicated by the name of the facility corresponding to the future life scene in the first correspondence information. The product with the calculated probability is calculated as the probability that the user is located in the facility type indicated by the name of the facility located after a predetermined time. Then, if the probability that the user is located in the facility type indicated by the estimated facility name is equal to or greater than a predetermined threshold value, the first position estimation unit 216 uses the probability as an estimation result.

On the other hand, if the probability that the user is located in the facility type indicated by the estimated facility name is less than a predetermined threshold value, the first position estimating unit 216 does not use the facility name as the estimation result. Next, the first position estimating unit 216 estimates the category of the facility that includes the name of the facility where the user is located after a predetermined time. Then, the product of the probability that the second estimated information is the next scene corresponding to the current life scene and the probability that the first corresponding information is located in the facility type indicated by the category of the facility corresponding to the future life scene is used. The probability that the person is located in the facility type indicated by the category of the facility located after a predetermined time is calculated, and the probability is used as the estimation result.

For example, it is assumed that the first position estimation unit 216 sets the threshold for the probability of the facility type where the user is located to 0.5, and the second scene estimation unit 115 estimates the future life scene as “dinner”. In this case, referring to the first correspondence information of FIG. 12, in the most detailed facility type about “dinner”, that is, the name of the facility, the probability 4/10 of using the A set restaurant is the mode value. This is not above the threshold of 0.5. Therefore, this estimation result is not used, and if the next facility type, that is, the facility category is referred to, the probability of using a set restaurant is 5/10, which is adopted because the threshold is 0.5 or more. That is, it is estimated that this user will transition to the “dinner” life scene after a predetermined time (for example, 30 minutes here), and the user is likely to stay at that time. The degree of certainty is insufficient to estimate, but it can be estimated that there is a 50% possibility of being located in the type of “canteen”. In addition, it can be estimated that “Ramen restaurant” which is another facility category is 30%, and “Family restaurant” is 20%.

Next, the position estimation process by the mobile terminal 200 according to the present embodiment configured as described above will be described. FIG. 13 is a flowchart of a position estimation process performed by the mobile terminal according to the second embodiment.

First, since the process from the current position measurement to the estimation of the future life scene (steps S51 to 55) is the same as the process (steps S11 to S15) in FIG. 9 of the first embodiment, a description thereof will be omitted.

Next, the first position estimation unit 216 estimates the name of the facility where the user is located after a predetermined time based on the future life scene based on the first correspondence information, and the user enters the facility type indicated by the name of the facility. The probability of being located is calculated (step S56). Then, the first position estimating unit 216 determines whether or not the calculated probability is equal to or greater than a predetermined threshold (step S57). If it is equal to or greater than the predetermined threshold (step S57: Yes), the output unit 118 outputs the estimated facility name and the calculated probability to the display unit as an estimation result (step S58).

On the other hand, when it is not more than a predetermined threshold (Step S57: No), the 1st position estimating part 216 presumes the category of the facility where a user is located after predetermined time based on the future life scene by the 1st correspondence information. Then, the probability that the user is located in the facility type indicated by the category of the facility is calculated (step S59). Then, the output unit 118 outputs the estimated facility category and the calculated probability to the display unit as an estimation result (step S60).

As described above, in the mobile terminal 200 according to the second embodiment, the current life scene of the user is estimated from the current position, the current operation state of the user, and the current time, and the estimated current life scene is obtained. Based on this, a future life scene after a predetermined time is estimated. Then, the mobile terminal 200 estimates the name of the facility or the facility category where the user is located after a predetermined time based on the estimated future life scene, and uses it for the facility type indicated by the estimated facility name or the facility category. The probability that the person is located is calculated, and these are output to the display unit. In this way, by using the current life scene of the user, the name of the facility where the user is likely to be located after a predetermined time from the continuity of the life scene without being estimated from the continuity of the user's position. Or since the category of the facility can be estimated, the name of the facility of the user or the category of the facility after a predetermined time can be estimated even at a place visited for the first time. Further, by using the first correspondence information in which the life scene and the facility type are hierarchically associated, it is possible to estimate the position of the user in more detail.

(Embodiment 3)
In the first and second embodiments, the future life scene is estimated from the user's current life scene, and the facility type where the user is located after a predetermined time is estimated based on the future life scene. On the other hand, in the present embodiment, a technique for further storing a user's movement history and estimating a position where the user is located after a predetermined time based on a user's movement pattern extracted from the stored movement history. Is added. Below, the structure which added the said technique to Embodiment 2 is demonstrated.

FIG. 14 is a block diagram of an example of a functional configuration of the mobile terminal according to the third embodiment. As illustrated in FIG. 14, the mobile terminal 300 includes a positioning unit 111, a detection unit 112, a first scene estimation unit 113, a storage unit 114, a second scene estimation unit 115, and a first position estimation unit 216. The scene update unit 117, the output unit 118, the extraction unit 321, the second position estimation unit 322, the determination unit 323, the first estimation storage unit 131, the second estimation storage unit 132, and the second correspondence memory Unit 141, first correspondence storage unit 242, history storage unit 351, and pattern storage unit 352. Note that the portable terminal 300 according to the third embodiment has a hardware configuration such as a CPU (not shown), a storage medium such as an HDD or a memory, as in the first and second embodiments. Here, since the same reference numerals as those in the first embodiment are the same as those in the first embodiment, the description thereof is omitted. The external configuration of the mobile terminal 300 according to the third embodiment is the same as that of the first embodiment.

The mobile terminal 300 according to the third embodiment stores the movement history of the current position measured by the positioning unit 111 in addition to the estimation of the future position based on the life scene transition in the mobile terminal 200 of the second embodiment. The future position is estimated based on the movement history, and the estimated future position is corrected.

The history storage unit 351 is a storage medium such as an HDD or a memory that stores a movement history in which the current position measured by the positioning unit 111 and the current time are recorded over time. FIG. 15 is a diagram illustrating an example of the movement history. As shown in FIG. 15, the movement history includes the current position (position indicated by latitude and longitude), the label for identifying the current position, and the name of the facility at the current position at the date and time when the current position is measured. Correspondingly recorded.

The pattern storage unit 352 stores a movement pattern in which the current position measured by the positioning unit 111 is associated with the probability that the user is located at the other position when moving from the current position to the other position. A storage medium such as an HDD or a memory. FIG. 16 is a diagram illustrating an example of the movement pattern. As shown in FIG. 16, the movement pattern includes “position 01”, “position 02”, etc., which are labels of the current position. "Position 02" and the like are associated with each other, and the probability "23/46" and "23/25" that the user is located at each next position is associated.

The extraction unit 321 generates a movement history from the current position and the current time measured by the positioning unit 111 and stores the movement history in the history storage unit 351. Then, a movement pattern is extracted based on the temporal transition of the current position in the stored movement history, and the extracted movement pattern is stored in the pattern storage unit 352. Specifically, the extraction unit 321 registers the location as a stay location when the user is present within a predetermined radius range for a predetermined time or more. And the extraction part 321 shall regard the passage as one stay, when a user passes the location registered later. For example, the predetermined radius is set to 100 m and the predetermined time is set to 5 minutes. And the extraction part 321 calculates the probability that the user was located in the next place by calculating the frequency of moving to another stay place after being located in each stay place, and the movement pattern is calculated. Extract.

The second position estimation unit 322 calculates, from the movement pattern, the probability that the user who owns the mobile terminal 300 is located at the next position corresponding to the current position after a predetermined time. For example, when the user is currently staying at the current position “position 04” from the movement pattern, the second position estimation unit 322 has a probability of 13/18 at the next position “position 03” after a predetermined time. It is presumed that it will move, and it is assumed that it will move to the next position “position 05” with a probability of 5/18. In the present embodiment, an estimation method for extracting a movement pattern from a movement history is described. However, a method using a Bayesian network or a hidden Markov model may be used.

The determination unit 323 determines the next position where the user is located after a predetermined time estimated by the second position estimation unit 322 based on the movement pattern, and the location of the user after the predetermined time estimated by the first position estimation unit 216. The final estimation result is determined by comparing with the facility type. That is, the determination unit 323 determines the next position estimated by the second position estimation unit 322 when the probability of being located at the next position estimated by the second position estimation unit 322 is equal to or greater than a predetermined threshold. The calculated probability is used as the estimation result. On the other hand, when the probability of being located at the next position estimated by the second position estimation unit 322 is less than a predetermined threshold, the facility type estimated by the first position estimation unit 216 and the calculated probability To do.

For example, assuming that a predetermined threshold value is 0.3, the determination unit 323 estimates the second position estimation unit 322 when the probability calculated by the second position estimation unit 322 is 0.3 or more. The determined next position is determined as a final estimation result and is output by the output unit 118. On the other hand, when the probability calculated by the second position estimation unit 322 is less than 0.3, the determination unit 323 determines the facility type estimated by the first position estimation unit 216 as the final estimation result. And output by the output unit 118.

With this configuration, when the user has visited many places in the past, an estimation using the movement history (estimation by the second position estimation unit 322) is output, and the user visited for the first time. When the user is in a place, the estimation based on the life scene described in Embodiment 2 (estimation by the first position estimation unit 216) can be output.

Next, position estimation processing by the mobile terminal 300 according to the present embodiment configured as described above will be described. FIG. 17 is a flowchart of a position estimation process performed by the mobile terminal according to the third embodiment. It is assumed that portable terminal 300 performs the position estimation process (FIG. 13) in the second embodiment in parallel with the following process.

First, the positioning unit 111 measures the current position of the mobile terminal 300 (step S71). Then, the extraction unit 321 generates a movement history from the current position and the current time and stores the movement history in the history storage unit 351 (step S72).

Next, the extraction unit 321 extracts a movement pattern based on the stored movement history, and stores the extracted movement pattern in the pattern storage unit 352 (step S73). Next, the second position estimation unit 322 calculates the probability that the user is located at the next position after a predetermined time based on the movement pattern (step S74).

And the determination part 323 judges whether the probability calculated by the 2nd position estimation part 322 is more than a predetermined threshold value (step S75). When it is equal to or greater than the predetermined threshold (step S75: Yes), the determination unit 323 determines the next position estimated by the second position estimation unit 322 and the calculated probability as an estimation result, and the output unit 118 The estimation result is output to the display unit (step S77). On the other hand, when it is not equal to or greater than the predetermined threshold (step S75: No), the determination unit 323 determines the facility type estimated by the first position estimation unit 216 and the calculated probability as an estimation result, and outputs the output unit 118. Outputs the estimation result to the display unit (step S76).

As described above, in the mobile terminal 300 according to the third embodiment, the facility type estimated from the current life scene of the user, the probability that the user determines for the facility type, and the movement history of the user actually moved When the next position estimated by the above and the probability that the user is located at the next position are calculated, an estimation result suitable for the user's situation can be output to the display unit. Accordingly, as in the second embodiment, the user's facility type after a predetermined time can be estimated even at a place visited for the first time, and the position can be estimated based on the movement history.

Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the above embodiments. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined.

As described above, the position estimation apparatus according to the present invention is useful for estimating a position where a user is located after a predetermined time, and in particular, the user is located after a predetermined time at a place where the user has visited for the first time. Suitable for estimating position.

100, 200, 300 Portable terminal 101 Positioning sensor 102 Body motion sensor 111 Positioning unit 112 Detection unit 113 First scene estimation unit 114 Storage unit 115 Second scene estimation unit 116, 216 First position estimation unit 117 Scene update unit 118 Output unit 118 display unit 131 first estimation storage unit 132 second estimation storage unit 141 second correspondence storage unit 142, 242 first correspondence storage unit 321 extraction unit 322 second position estimation unit 323 determination unit 351 history storage unit 352 pattern storage unit

Claims (10)

First estimation information storing a life scene, which is an action scene in a user's life, a time zone of the life scene, a position of the life scene, and a user's operation state in the life scene is stored. An estimated storage unit;
A first correspondence storage unit that stores first correspondence information in which the life scene is associated with a facility type in which the user is located in the life scene;
A second estimation storage unit that stores second estimation information including the life scene and a next scene representing a scene of an operation performed by the user next to the life scene;
A positioning unit that measures the current position;
A detection unit for detecting the current operating state of the user;
A first scene estimation unit that estimates a current life scene of a user from the first estimation information using the life scene corresponding to the current position, the current operation state, and the current time;
A second scene estimation unit for estimating a future life scene, which is the life scene after a predetermined time of the user, from the second estimation information using the next scene corresponding to the current life scene;
From the first correspondence information, using the facility type corresponding to the future life scene, a first position estimation unit for estimating the facility type where a user is located after a predetermined time;
A position estimation apparatus comprising: A second correspondence storage unit that stores second correspondence information in which a facility type indicating a type of facility is associated with a position of the facility type;
The first position estimating unit further uses the position of the facility type corresponding to the facility type where the user is located after a predetermined time from the second correspondence information to determine the position where the user is located after the predetermined time. The position estimation apparatus according to claim 1, wherein estimation is performed. The apparatus further comprises a storage unit that stores the first correspondence information in which the current life scene is associated with the facility type corresponding to the current position in the second correspondence information in the first correspondence storage unit. The position estimation apparatus according to claim 2. The second estimation information includes the life scene and the next scene accompanied by a probability that a user's action is the next scene,
The first correspondence information further associates a probability that a user is located in the facility type corresponding to the life scene,
The first position estimating unit further sets the probability that the second scene information corresponds to the current life scene in the second estimation information and the facility type corresponding to the future life scene in the first correspondence information. The position estimation apparatus according to claim 3, wherein a product of the location probability is calculated as a probability that a user is located in the estimated facility type. The second correspondence information further associates the probability that the user is located at the location of the facility type corresponding to the facility type,
The first position estimating unit further sets the probability that the second scene information corresponds to the current life scene in the second estimation information and the facility type corresponding to the future life scene in the first correspondence information. Probability that the user is located at the estimated position, the product of the probability of being located and the probability that the user is located at the location of the facility type corresponding to the facility type located after a predetermined time in the second correspondence information The position estimation device according to claim 4, wherein 6. The position estimation apparatus according to claim 5, further comprising a scene update unit that updates a probability of the next scene in the second estimation information based on a user's action. The facility type includes a facility category and a facility name included in the facility category,
The first correspondence information associates the life scene, the facility type, and the probability of being located in the facility type indicated by the name of the facility,
The first position estimator further includes a probability of becoming the next scene corresponding to the current life scene in the second estimation information, and a name of the facility corresponding to the future life scene in the first correspondence information. The position according to claim 6, wherein the product of the probability of being located in the facility type indicated by is calculated as the probability that the user is located in the facility type indicated by the name of the facility located after a predetermined time. Estimating device. The first correspondence information further associates the life scene with the probability of being located in the facility type indicated by the facility category,
The first position estimation unit further corresponds to the current life scene in the second estimation information when the probability of being located in the facility type indicated by the calculated name of the facility is less than a predetermined threshold. The product of the probability of becoming the next scene and the probability of being located in the facility type indicated by the facility category corresponding to the future life scene in the first correspondence information is the product of the facility where the user is located after a predetermined time. The position estimation apparatus according to claim 7, wherein the position estimation apparatus calculates the probability of being located in the facility type indicated by the category. A pattern storage unit that stores a movement pattern in which the current position is associated with a probability that a user who has moved from the current position to another position is located at the other position;
A second position estimation unit that calculates a probability that the user is located in the other position corresponding to the current position after a predetermined time from the movement pattern;
When the probability of being located in the other position is equal to or greater than a predetermined threshold, the probability calculated by the second position estimation unit is an estimation result, and the probability of being located in the other position is less than a predetermined threshold, The position estimation apparatus according to claim 8, further comprising: a determination unit that uses the probability calculated by the first position estimation unit as an estimation result. A history storage unit for storing a movement history in which the current position and the current time are recorded over time;
The apparatus further comprises: an extraction unit that extracts the movement pattern based on temporal transition of the current position in the movement history and stores the extracted movement pattern in the pattern storage unit. The position estimation apparatus according to 9.

Images