JP6681657B2 - Travel status determination device, travel status determination method and program - Google Patents

Description

  The present invention relates to a travel status determination device, a travel status determination method, and a program.

  Conventionally, there has been proposed a technology aiming at automatically estimating a user's action purpose even if the user does not input the action purpose, and presenting guidance information useful to the user (for example, patents). Reference 1). According to the technique, in the navigation device, the action purpose of the user is estimated based on the information on the person, the information on the object, etc. existing in the vehicle.

  There is also proposed a technique of predicting a moving purpose by using a captured image and a decoration state of a vehicle occupant recognized from the image (for example, Patent Document 2). According to the technology, it is possible to accurately estimate a destination of a less habitful action.

  Further, modeling information of the co-occurrence relationship is generated using the staying place and staying time of the user and the scheduler data that defines the action schedule of the user, and the user's action is predicted by using the modeling information. The technique called is also proposed (for example, patent document 3). The technology is said to be able to predict unusual behavior of users.

Japanese Patent No. 4925808 Japanese Patent No. 5286952 Japanese Patent No. 5433472

  Conventionally, a technique of estimating a user's destination has been proposed. Here, in order to estimate the purpose of movement, in addition to the past movement history, various information such as an image of an occupant and scheduler data defining a user's action schedule has been required. Therefore, an object of the present invention is to provide a technique for estimating the situation where the user is placed and the purpose of the movement based on the history information regarding the movement of the mobile body.

  A movement status determination device according to the present invention includes a movement history generation unit that generates a movement history including a destination of a moving body based on position information indicating a position of the moving body, and a history of the moving body based on the movement history. The model generation unit that aggregates the number of arrivals at each destination related to the movement for each arrival time zone, and the destination of the moving body is the destination with the largest number of arrivals in the past in the estimated arrival time zone. In this case, the mobile body has an addictiveness determination unit that determines that the movement of the mobile body is a habitual action, and an information output unit that outputs information to the mobile body according to the determination of the habitality determination unit.

  With such a configuration, it is possible to determine whether or not the movement of the moving body is a habitual action based on the movement history. In addition, based on the determination, it is possible to estimate that it is a situation in which there is not enough time for habitual behavior, for example, and it is possible to output information according to this. That is, the situation where the user is placed and the purpose of the movement can be estimated based on the history information regarding the movement of the moving body.

Further, the destination may be a place estimated based on the movement history and the movement route from the departure place of the moving body to the present place. By doing so, it becomes possible to estimate the situation where the user is placed and the purpose of the movement without receiving the designation of the destination from the user.

  Further, when the starting point and the destination of the moving body acquired by the position information acquiring unit are within a predetermined range from the position of the user's home related to the moving body, or the moving body acquired by the position information acquiring unit. Of the departure point, and the destination, if the passing frequency in the movement history is included in a region equal to or more than a threshold value, further has a living range outside determination unit that determines that the movement is within the living range, the information output unit, Information may be output to the mobile body in accordance with the determination made by the habit determination unit and the determination made by the outside-living-region determination unit. By doing this, in addition to the above-mentioned criteria of whether or not the behavior is a habitual behavior, the criteria of whether or not the movement is within the living area where the user has a sense of land are combined, and the situation in which the user is placed You will be able to estimate the purpose of the movement.

  Further, the position of the home of the user relating to the moving body may be a place estimated based on the movement history. By doing so, it becomes possible to estimate the situation where the user is placed and the purpose of the movement without the user having to set the user's home.

  The movement history may represent a transition of position information of the moving body based on a mesh that divides the area into a mesh shape having a predetermined position and size. In this way, the movement history can be represented without using the information of the road that changes with time, so that the cost for maintaining and managing the map information can be reduced.

  The contents described in the means for solving the problems can be combined as much as possible without departing from the problems and technical ideas of the present invention. The contents of the means for solving the problems can be provided as a device such as a computer or a system including a plurality of devices, a method executed by the computer, or a program executed by the computer. A recording medium holding the program may be provided.

  According to the present invention, it is possible to provide a technique of estimating a situation where a user is placed and a purpose of movement based on history information regarding movement of a mobile body.

It is a figure which shows an example of a system structure. It is a functional block diagram which shows an example of a movement condition determination apparatus. It is a figure for demonstrating a local mesh code. It is a figure for demonstrating the number of times of arrival at the destination totaled for each day of the week and time. It is a figure for demonstrating determination of a moving condition. It is an apparatus block diagram which shows an example of a computer. It is a processing flow figure showing an example of model creation processing. It is a figure which shows an example of position information. It is a figure which shows an example of a movement history. It is a processing flow figure showing an example of situation discriminating processing. It is a schematic diagram for explaining a mesh whose passing frequency is equal to or higher than a threshold. It is a figure which shows an example of a conditional probability table. It is a figure for demonstrating the digraph between meshes. It is a figure which shows an example of a conditional probability table at the time of movement. It is a figure for demonstrating a Bayes update. It is an example of a default destination arrival probability.

  Embodiments of the present invention will be described below with reference to the drawings. The following embodiments are examples of the present invention, and the present invention is not limited to the following configurations.

<System configuration>
FIG. 1 is a diagram showing the configuration of the system according to the present embodiment. The present system includes a movement status determination device 1, a network 2, and a moving body 3 (3a, 3b ...). The mobile unit 3 includes a sensor such as a GPS (Global Positioning System) receiver that can acquire position information. In the present embodiment, the moving body 3 will be described as transmitting position information. Specifically, the moving body 3 has a vehicle 3a such as a passenger car or a two-wheeled vehicle having a navigation device, and a mobile device 3b such as a mobile phone, a smartphone, a tablet (slate PC (Personal Computer)), or a notebook PC. Users who go out. Further, predetermined application software is installed in the navigation device and the mobile device, and the application software continuously transmits the position information to the movement situation determination device 1 via the network 2. On the other hand, the movement status determination device 1 determines whether or not the movement is in an area where there is a sense of land based on the current position and destination of the moving body 3, a model of a movement pattern generated from past movement history, and the like (rule). (Also referred to as sex), a movement status such as whether or not there is a movement is determined, and information corresponding to the determination is output.

<Functional configuration>
FIG. 2 is a functional block diagram showing an example of the movement status determination device 1. The movement status determination device 1 includes a storage unit 11, a position information acquisition unit 12, a movement history generation unit 13, a model generation unit 14, a predicted destination acquisition unit 15, a home position acquisition unit 16, and a living area outside. The determination unit 17, the habit determination unit 18, and the information output unit 19 are included.

  The storage unit 11 is a so-called main storage device or auxiliary storage device, and temporarily or permanently holds information used in this embodiment. The position information acquisition unit 12 continuously acquires position information including latitude and longitude from the mobile unit 3 via the network 2 and stores it in the storage unit 11. In addition, the movement history generation unit 13 generates a movement history related to one movement from a plurality of position information stored in the storage unit 11 and stores the movement history as one unit indicating the movement from the departure place to the destination. It is stored in the section 11. It should be noted that the start and end of the movement are determined, for example, from the engine start of the vehicle to the engine stop, or from the stay in substantially the same place for a predetermined period or longer to the stay in another place for a predetermined period or longer. . In addition, in this embodiment, the destination reached in the past movement is also referred to as a “visit”.

  In the present embodiment, position information including latitude and longitude is converted into, for example, a regional mesh code, and a movement history representing transition of the regional mesh code is generated. FIG. 3 is a diagram for explaining a regional mesh code. In this embodiment, a regional mesh code defined in JIS X0410 of Japanese Industrial Standard is used. The regional mesh code includes a primary mesh (also referred to as "first regional partition"), a secondary mesh (also referred to as "secondary regional partition", "integrated regional mesh"), and a third mesh ("second regional partition"). The size of each stage of standard area meshes such as "tertiary area divisions" and "reference area meshes") and divided area meshes such as half area meshes, quarter area meshes and eighth area meshes. The position is expressed by the mesh of.

The primary mesh is a region divided into approximately rectangular areas each having a side length of about 80 km. The primary mesh code is the upper two digits of latitude and the lower two digits of longitude. It is represented by a 4-digit number consisting of digits. The secondary mesh is obtained by dividing the primary mesh into 8 parts in the east-west direction and the north-south direction, and the secondary mesh code is the upper 1-digit number indicating the latitude direction and the lower 1-digit number indicating the longitude direction. It is expressed in a format in which a two-digit number consisting of the number is connected to the primary mesh code with "-(hyphen)". The 3rd mesh is obtained by dividing the 2nd mesh into 10 parts in the east-west direction and the north-south direction, and the 3rd mesh code is the 1st digit of the latitude direction and the 1st digit of the longitude direction. It is expressed in a format in which a two-digit number consisting of the number is connected to the primary mesh code and the secondary mesh code. The 1/2 area mesh is obtained by halving the third mesh in the east-west direction and the north-south direction, and the 1/2 area mesh code is 1 in the southwest area, 2 in the southeast area, and northwest. It is expressed in a form in which the regions of 3 and the region of the northeast are 4 and are connected to the primary to tertiary mesh codes. The quarter area mesh is obtained by equally dividing the half area mesh in the east-west direction and the north-south direction, and the quarter area mesh code is 1 in the southwest area and 2 in the southeast area. , The northwestern region is 3, and the northeastern region is 4, which are connected to the primary to half regional mesh codes. The 1/8 area mesh is a 1/4 area mesh divided into two parts in the east-west direction and north-south direction. The 1/8 area mesh code is 1 in the southwest area and 2 in the southeast area. , The northwestern region is 3, and the northeastern region is 4, which is expressed in the form of being connected to the primary to quarter region mesh codes.

  Since the conversion process from the position information including latitude and longitude to the regional mesh code can be performed using an existing algorithm, the details will be omitted. In addition, in the present embodiment, for convenience, the primary mesh side is relatively referred to as an upper mesh, and the ⅛ area mesh side is relatively referred to as a lower mesh. The mesh is not limited to the existing area mesh code, and a mesh that divides the area into a mesh having predetermined positions and sizes may be uniquely defined and used.

Unlike the method that requires map information, such as the case where movement history is represented by a series of location information with intersections as nodes, the use of regional mesh codes reduces the cost for updating and managing map information. be able to. In addition, the mesh code is associated with and saved with commercial facilities, tourist spots, and other landmarks (for example, POI (Point Of Interest)), and the mesh code series is associated with traffic information such as traffic jams and construction. By storing the information in advance, it is possible to output appropriate information from the movement status determination device 1 to the moving body 3. As described above, the regional mesh code can represent the position information with a plurality of levels of detail. By changing the degree of detail according to the movement distance of the moving body 3 and the processing load of the movement status determination device 1, it is possible to improve the accuracy of status determination and the processing speed.

  In addition, the model generation unit 14 generates a history model used in a determination process described below from the movement history stored in the storage unit 11 and stores the history model in the storage unit 11. The history model according to the present embodiment refers to statistical information that is generated from the movement history and that represents the habitual tendency. For example, the number of times of arrival at each destination (that is, the visited place) is totaled for each day of the week and each time period. FIG. 4 is a diagram for explaining the number of times of arrival at the destination, which is calculated for each day of the week and each time period. In the example of FIG. 4, as the aggregation result for the past 8 o'clock on Thursday, 43 times for the company, 6 times for the barber, and 3 times for the gym are stored. Although the destinations are aggregated for each landmark in FIG. 4, the destinations may be aggregated for each mesh. The model generation unit 14 may also obtain an index value for habit determination. For example, it is determined whether or not the ratio between the destination having the largest number of visits and the destination having the second or less visit at a certain time zone on a certain day is equal to or greater than a predetermined threshold. Then, if there is a difference equal to or more than a predetermined threshold value, it is determined in the determination process described later that the user is habitually moving toward the destination with the highest number of visits during the time period of the day of the week. .

The destination acquisition unit 15 reads out information indicating the destination of the mobile unit 3 from the storage unit 11. The destination of the moving body 3 may be a position input by the user (driver) to the navigation device, or may be predicted by the movement status determination device 1 or the like based on the movement from the movement start of the moving body 3 to the current position. It may be a position. Further, the home position acquisition unit 16 reads out the home position of the user of the mobile unit 3 from the storage unit 11. As for the home position, the position input by the user to the navigation device may be read out, or the home position estimated based on the movement history may be read out.

  The outside-living-area determining unit 17 determines whether or not the current movement is movement within the living area of the user of the moving body 3 based on the starting point and the destination of the moving body 3 that is moving. For example, if both the departure place and the destination are within a predetermined distance from the position of the home, it is determined to be within the living area. In addition, the habituation determination unit 18 determines whether or not the current movement is habitual for the user of the moving body 3, based on the destination of the moving body 3 and the day and time of day when the moving body is moving. Determine whether. A habitual movement is a daily rule that is presumed to have to be moved in time for a predetermined time, such as movement to a fixed place on the same day of the week and the same time zone. Behavior.

  FIG. 5 is a diagram for explaining the determination of the movement status. In the present embodiment, it is classified into four movement situations (also referred to as scenes) that represent the situation in which the user of the mobile unit 3 is placed and the purpose of the movement, depending on whether the user is outside the living area or habitual. Specifically, the situation is classified into daily life and extraordinary life depending on the presence or absence of habit. In addition, the situation is classified into Home and Away according to the living area. When these situations are combined, they are classified into four movement situations: extraordinary and home, extraordinary and away, everyday and home, everyday and away.

  The information output unit 19 determines whether or not the user of the mobile body 3 is moving in an area with a sense of land, depending on conditions such as whether or not the movement is within the living area and whether or not the movement is habitual, Also, the movement status such as whether or not it is a routine routine movement is determined, and information according to the situation is output to the mobile body 3 via the network 2.

<Device configuration>
FIG. 6 is a device configuration diagram showing an example of a computer. The movement status determination device 1, the navigation device of the mobile unit 3 (not shown), the mobile terminal, and the like are computers, for example, as shown in FIG. A computer 1000 shown in FIG. 6 includes a CPU (Central Processing Unit) 1001, a main storage device 1002, an auxiliary storage device (external storage device) 1003, and a communication IF.
An interface 1004, an input / output IF (Interface) 1005, a drive device 1006, and a communication bus 1007 are provided. The CPU 1001 executes the processing according to the present embodiment by executing the program. The main storage device 1002 caches programs and data read by the CPU 1001 and expands the work area of the CPU. The main storage device is specifically a RAM (Random Access Memory), a ROM (Read Only Memory), or the like. The auxiliary storage device 1003 stores a program executed by the CPU 1001, position information, and the like. The auxiliary storage device 1003 is, specifically, an HDD (Hard-disk Drive) or an S.
These are SD (Solid State Drive), eMMC (embedded Multi-Media Card), flash memory and the like. The main storage device 1002 and the auxiliary storage device 1003 work as the storage unit 11 and the like of the movement status determination device 1. The communication IF 1004 transmits / receives data to / from another computer. The movement status determination device 1 is connected to the network 2 via the communication IF 1004. The communication IF 1004 is specifically a wired or wireless network card or the like. The input / output IF 1005 is connected to the input / output device, receives an input from the user, and outputs information to the user. The input / output device is specifically a keyboard, a mouse, a display, a touch panel, or the like. The drive device 1006 reads data recorded in a storage medium such as a magnetic disk, a magneto-optical disk, an optical disk, or writes data in the storage medium. The above-described components are connected by the communication bus 1007. A plurality of these constituent elements may be provided, or some of the constituent elements (for example, the drive device 1006) may not be provided. The input / output device may be integrated with the computer. Further, the program executed in the present embodiment is provided via a portable storage medium readable by the drive device 1006, a portable auxiliary storage device 1003 such as a flash memory, a communication IF 1004, and the like. You may Then, the CPU 1001 executes the program to cause a computer as shown in FIG. 6 to operate as the movement situation determination device 1 and the like shown in FIG.
<Model creation processing>
Next, details of the processing of the movement status determination device 1 will be described. The movement status determination device 1 continuously receives the position information from the moving body 3, stores the movement history, and generates the prediction model.

  FIG. 7 is a process flowchart showing an example of the model creating process. The position information acquisition unit 12 of the movement status determination device 1 acquires position information from the mobile unit 3 via the network 2 at an arbitrary timing (FIG. 7: S1). First, the mobile unit 3 measures position information indicating its own position by latitude and longitude, for example, by a GPS receiver. Then, the moving body 3 transmits the position information to the movement status determination device 1 at a predetermined time, a predetermined distance, or the like. The position information may be so-called probe information. On the other hand, the position information acquisition unit 12 of the movement status determination device 1 stores the received position information in the storage unit 11.

  FIG. 8 is a diagram showing an example of the position information stored in the storage unit 11. In the example of FIG. 8, as the position information of a certain moving body at a certain time, a value corresponding to each item such as the moving body ID, date and time, latitude, longitude, direction, and speed is stored in one record. The direction item indicates the traveling direction of the moving body, and the value indicating the traveling direction is represented by, for example, an azimuth angle where true north is 0 degree or 12:00.

  Then, the position information acquisition unit 12 determines whether the mobile unit 3 has arrived at the destination (that is, the visited place) (S2). The position information acquisition unit 12 may determine that the mobile unit 3 has arrived at the destination when the position information is not transmitted from the mobile unit 3 for a predetermined time or longer, or the mobile unit 3 arrives at the destination. You may make it transmit the information to the effect. When it is determined that the engine stop of the vehicle has arrived at the destination, when the power-off of the vehicle-mounted device is instructed, the information indicating the arrival at the destination and the position information of the point are transmitted to the movement status determination device 1. After that, you may shut down. Further, when the control in conjunction with the power supply of the engine or the in-vehicle device cannot be performed, when the power of the in-vehicle device is turned on next time, the positional information of the point and the date and time immediately before the previous shutdown are sent to the movement status determination device 1. You may make it transmit. When it is determined that the vehicle has not arrived at the destination (S2: NO), the process returns to S1, and the position information acquisition unit 12 sequentially stores the position information transmitted by the mobile unit 3 in the storage unit 11.

On the other hand, when it is determined that the destination has been reached (S2: YES), the movement history generation unit 13 of the movement status determination device 1 generates a movement history in which the unit from the start of movement to the arrival of the destination is one unit ( S3). In this step, for example, one record of the movement history as shown in FIG. 9 is generated. The movement history in FIG. 9 includes values corresponding to items such as a moving body ID, history ID, departure information, arrival information, traveling time, passing point, and weather. The departure information and the arrival information each include a value corresponding to each item of mesh, latitude, longitude, POI, and date and time. The identification information of the moving body is registered in the item of the moving body ID. In the item of history ID, the identification information of the movement history indicating from the start of movement to the arrival at the destination is registered. In addition, the POI information updating unit (not shown) of the movement status determination device 1 stores the position information represented by the regional mesh and the information of the facility or the like existing at the position in the storage unit 11 in association with each other. You may In addition, the weather information acquisition unit (not shown) may acquire information indicating weather for each region and for each past time zone from, for example, an information distribution server on the Internet and store the acquired information in the storage unit 11. Good. Then, the movement history generation unit 13 reads out the POIs respectively associated with the position information of the departure place and the arrival place and the weather information of the departure time at the departure place and stores them in the movement history. The weather information is, for example, weather classification such as fine, cloudy, rain, snow, unknown, and the maximum temperature and the minimum temperature. Further, the travel time item holds the required time obtained by subtracting the date and time of departure from the date and time of arrival. The passage point item holds a series of one or more points passing from the departure point to the arrival point.

After that, the model generation unit 14 of the movement status determination device 1 generates a history model (S4). The history model is statistical information indicating a habitual tendency in the movement history, and is used as an index in the movement situation determination process described later. In this step, the number of times of arrival at the destination, which is calculated for each day of the week and each time period, is recorded in the statistical information as shown in FIG. That is, the model generation unit 14 reads the movement history information generated in S3, and determines the number of arrivals of the corresponding destination (POI of arrival information) of the cell of the corresponding day of the week and time zone based on the date and time of the arrival information. Increment 1. In addition, in this step, the number of visits between the destination with the highest number of visits in a certain time period on a certain day of the week and the other destinations (the number of visits is the second or less) are used as an index used in the habit determination. The ratio should be calculated. For example, the index shown in the following formula 1 is obtained for each day of the week and each time zone, and when the index is equal to or more than a predetermined threshold value, the destination with the highest number of visits is associated with the day of the week and the time zone. Keep it as a "customary destination".

  With that, the model generation process is completed. Each time the moving body 3 starts moving, the model generation process of FIG. 7 is started and the history model is updated.

<Situation determination processing>
Next, the situation determination process using the history model will be described. The situation determination process is executed while the moving body 3 is moving. That is, in the model creation process described above, for example, the process is performed in parallel with the position information acquisition process.

  FIG. 10 is a process flow chart showing an example of the situation determination process. The destination acquisition unit 15 of the movement status determination device 1 acquires the information of the destination from the storage unit 11 (FIG. 10: S11). The information on the destination is stored in the storage unit 11 in the form of, for example, a landmark or a mesh. Further, the destination information may be based on the information input by the user to the navigation device or the like, or may be estimated by the movement situation determination device 1 or the navigation device or the like. The destination estimation processing will be described later.

  In addition, the home position acquisition unit 16 of the movement status determination device 1 acquires home information from the storage unit 11 (S12). The home information is stored in the storage unit 11 in a mesh format, for example. The home position may also be based on the information input by the user to the navigation device or the like, or may be estimated by the movement status determination device 1 or the navigation device or the like. The home position estimation process will also be described later.

  After that, the outside-living-area determining unit 17 of the movement status determination device 1 determines whether or not the movement of the moving body 3 is within the living area of the user of the moving body 3 (S13). For example, the outside-inside-living-area determining unit 17 determines that both the departure place related to the current movement of the moving body 3 acquired by the position information acquiring unit 12 and the destination acquired by the destination acquiring unit 15 are the home position acquiring unit. If 16 is within a predetermined distance from the acquired home position, it is determined to be within the living area. On the other hand, if either the departure place or the destination is not within the predetermined distance from the home, it is determined to be out of the living area. The term “within a predetermined distance” means that a threshold value is predetermined, for example, a linear distance from home is within 3 km.

  In S13, the outside-living-area determining unit 17 may determine, for example, a mesh or a section whose passing frequency is equal to or more than a predetermined threshold in the movement history as the inside-living area. FIG. 11 is a schematic diagram for explaining a mesh whose passing frequency is a threshold value or more. In the example of FIG. 11, for convenience, the meshes are denoted by reference numerals A to J in the vertical direction and 1 to 10 in the horizontal direction. It is also assumed that the mesh F5 has the user's home. Then, using the mesh related to the departure information of the movement history, the mesh related to the arrival information, and the mesh related to the passing points shown in FIG. The pattern is filled so that the number of dots increases. Further, it is assumed that the mesh whose passing count is equal to or greater than a predetermined threshold is shown in a thick frame as shown in FIG. Here, for convenience, a mesh group in which the number of passages is equal to or larger than a predetermined threshold (that is, a mesh surrounded by a thick frame) is also referred to as a mesh set. In the example of FIG. 11, the number of passages of the mesh around the home is relatively high, but even if the meshes are approximately equidistant from the home, the number of passages is uneven. By using such information, for example, when both the departure place and the destination are either of the mesh sets, it is possible to determine that the movement is within the living area. In addition, it may be determined that the movement is within the living area when a certain proportion or more of the meshes predicted to pass from the starting point to the destination is one of the mesh sets. Similarly, by using the mesh set, it is possible to determine that a place other than home and that is frequently visited is also within the living range (in other words, a region with a sense of land).

  Then, the habituation determination unit 18 of the movement status determination device 1 determines whether or not the current movement of the moving body 3 is a movement with habit (S14). In this step, the presence or absence of addiction is determined using the history model and index created in the model creation process. For example, when the destination of the mobile unit 3 is the destination having the largest number of visits on the same day of the week and the same time zone in the past, it is determined that the movement is habitual. Furthermore, when the destination acquired in S11 is the customary destination calculated in S4, it may be determined that the travel is habitual. That is, it is determined that the travel is habitual if the destination has a high visit probability of a predetermined point or more on the day of the week and the time of day.

  After that, the information output unit 19 of the movement status determination device 1 outputs information according to the determination result of the movement status to the mobile unit 3 via the network 2 (S15). In this step, different information is output depending on the four types of movement statuses of extraordinary and home, extraordinary and away, routine and home, and everyday and away shown in FIG. The information is assumed to be stored in the storage unit 11 in association with the mesh, and for example, the information associated with the mesh that the moving body 3 is passing through or the mesh existing on the route to the destination is output.

  In the present embodiment, in the case of home, it is determined that the user is moving within an area where the user has a sense of land, that is, within a living area. Therefore, for example, local information such as the opening of a new store and recommended information closely related to life are provided. On the other hand, in the case of away, various information is output according to the purpose of the user's movement, such as tourist spot information, driving route proposals, traffic information such as danger points, and the like. In addition, in everyday cases, unusual suggestions are not given, and traffic information to the destination is guided so that the user will not be late for a predetermined time. On the other hand, in the case of extraordinary days, it is judged that there is a possibility that there is a time margin, and various information on the area is output, and suggestions regarding routes and destinations are made. Hereinafter, information output in four types of movement situations related to the combination will be illustrated.

<Output example 1>
In the case of extraordinary and away, various information on the area may be introduced. For example, it outputs information about military commanders related to the area related to history, historical episodes of tourist destinations, origins of place names, and the like. In addition, local transportation and vehicles may be introduced, or tourist sites such as ruins may be introduced. Such information may be output according to the preference estimated based on the movement history of the user (moving body), or a new proposal may be made regardless of the movement history. .

<Output example 2>
Similarly, in the case of extraordinary and away, music to be played may be suggested depending on the situation. Music, for example, suggests something about the area in transit. In addition, a statistical processing unit (not shown) receives the position information and the identification information of the music reproduced at the point from the user of the system according to the present embodiment, and reproduces it in the area where another user is moving. You may make it suggest music with a high ratio. At this time, music data may be distributed.

<Output example 3>
In the case of extraordinary and away, a tourist plan may be proposed. In this case, for example, the user's movement history is used to estimate preferences, family structure, and the like, and a tourist spot based on the estimation is proposed.

<Output example 4>
In addition, in the case of extraordinary and away, traffic information that prompts safe driving may be output. By calling attention based on statistical information such as accidents, driving on unknown land can be supported.

<Output example 5>
In the case of extraordinary and away, the movement status determination device 1 may propose a drive plan to an area where there is no visit history. The movement status determination device generates a plan of a route, a stop-by point, a rest point, or the like, for example, according to the above-described estimation of the preference or the like.

<Output example 6>
In addition, when it is determined that it is extraordinary and home, and when it is determined that the passenger departs earlier than the habitual movement, it may be suggested to stop at a coffee shop or a fast food restaurant. . In particular, when it is determined to be extraordinary, it is preferable to make a proposal in an interactive form with the user, because the behavior pattern is different from the normal behavior pattern.

<Output example 7>
In cases where it is difficult to estimate the situation, it is preferable to appropriately grasp the purpose of the movement interactively rather than making an inappropriate proposal. If you start moving at midnight, which is not recorded in the history, by asking the user to input the purpose, you can appropriately understand whether the family is suddenly ill or if you stay overnight in the car and go to the ski resort. .

  As described above, according to the present embodiment, it is possible to estimate the situation where the user is placed and the purpose of the movement based on the history information regarding the movement of the mobile body. Therefore, it becomes possible to output appropriate information according to the purpose and situation of the movement. In particular, if the position of the home or the destination is also estimated, it is possible to provide the information determined to be appropriate for a certain user's movement based on the history information without accepting the input of any information from the user. become able to. Next, the contents of the home estimation process and the destination estimation process will be illustrated.

<Home estimation process>
In the home estimation process, the history information shown in FIG. 9 is read out, and the number of arrivals of each arrival information for each mesh (that is, a visited place) is totaled. Then, the mesh that has reached the maximum number of times is determined to be the area where the home exists. This is because the user moves from his or her home, and it is considered that the home is visited most frequently. Note that the determination may be made by further adding a condition that the visit frequency is equal to or higher than a predetermined threshold value. By doing this, for example, when there is a second-rank mesh whose visit counts are equal, the judgment is suspended. According to such home estimation processing, the home position can be estimated without receiving information input from the user. Further, as in the present application, if the home position is held in units of meshes having an area instead of the pinpoint latitude and longitude, it becomes a system in consideration of privacy related to the user's address.

<Destination estimation processing>
In the destination estimation process according to the present embodiment, for example, a prediction model used in the prediction process at the time of departure and a prediction model used in the prediction process at the time of movement are generated in advance. Note that (1) Hidden Markov Model Method and (2) Bayesian Network Method are illustrated as the prediction model used for the prediction process at the time of departure. On the other hand, (3) similarity method and (4) Bayesian update method are exemplified as the prediction model used for the prediction processing during movement. Then, when the moving body 3 starts moving, the destination is estimated using the prediction model. Hereinafter, each method will be described.

(1) Hidden Markov Model Method In the Hidden Markov Model Method, a pattern of a series including a plurality of combinations of past starting points and destinations (that is, destinations) and a pattern of a series of transit points related to movement to reach the present location. Based on the similarity with, the destination of this movement is predicted stochastically. In the present embodiment, it is assumed that the destination is estimated based on the movement history from the starting point to the destination for a predetermined number of times (N) in the past (based on the so-called Nth floor Markov chain). That is, a combination of a starting point and a destination for a predetermined number of times in the past is used as a prediction model. In this way, the destination can be appropriately predicted when the movement of the moving body has the regularity that the destination moves along the predetermined route. The prediction model may be generated in consideration of conditions such as the day of the week, departure time, and weather. In this way, immediately after the moving body starts moving, the next destination can be estimated based on the destinations passed in the past.

以上のようにして集計した重み付きの訪問回数を正規化し、確率分布を求める。 (2) Bayesian network method In this method, a prediction model that is the probability of reaching each destination (that is, a destination) is calculated for each condition at the time of departure based on the accumulated movement history. For example, the transition probability to each destination is calculated for each combination of the departure place, the departure day of the week, and the departure time, which are the conditions at the time of departure. In detail, the following processing is performed.
(1-1) One newly created record is read out from the movement history, and the condition at the time of departure represented as a combination of the departure place, the departure day of the week, the departure time, etc. is extracted. The starting point and the destination are represented by, for example, a quarter area mesh or POI.
(1-2) The read records are classified into predetermined categories based on the extracted conditions at the time of departure. For example, the destinations and the departure times are classified into categories represented by a combination of conditions for each of six time zones from 0:00 to four hours and for each day of the week.
(1-3) A predetermined adjustment value u is added to the total number of visits to the destination stored in association with the cells of the conditional probability table (CPT) based on the above combinations. Note that generally u = 1 and 1 is added for each visit, but recent movement history may be weighted higher. For example, u is defined as the following Expression 2. Note that d is, for example, the number of days from the earliest date in the movement history to the date on which the record to be processed is generated.

The number of visits with weights collected as described above is normalized to obtain a probability distribution.

FIG. 12 is a diagram showing an example of the conditional probability table. Each cell of the conditional probability table corresponds to a category represented by a combination of the above conditions. Further, in each category, a probability distribution indicating the frequency of each destination is aggregated. Such a conditional probability table is generated for each departure place. In the destination estimation process described later, the probability distribution of the destination is read out based on the conditions at the time of departure, and the destination with the highest probability is set as the predicted destination. In the model generation process, if the departure day is Monday to Friday, it may be classified as a weekday, and if Saturday or Sunday, it may be classified as a holiday. The prediction model may be generated by further using the weather information and the like. By doing this, immediately after the moving body starts moving, the next destination based on the probability distribution that has been aggregated under the same starting point, the same day of the week, the same time zone, the same weather, etc. in the past. Can be estimated.

(3) Similarity method Next, a prediction model used for destination prediction during movement will be described. In the prediction process during movement, position information is represented by a directed graph (also referred to as a section) between adjacent meshes. In the similarity method, a score representing the likelihood of each destination is calculated based on the similarity of the sections existing on the route from the starting point to the current point.

  FIG. 13 is a diagram for explaining a directed graph between meshes. In the example of FIG. 13, for convenience, the reference numerals A to J are given to the vertical direction of the mesh and the symbols 1 to 10 are given to the horizontal direction. Further, FIG. 13 shows 100 meshes A1 to J10. Here, a series of position information representing a route when moving linearly from G2 to D8 will be described. First, the moving body moves from G2 to G3. A path is represented by a pair of adjacent meshes having such directions. After that, the moving body moves from G3 to F3, F3 to F4, F4 to F5, F5 to E5, E5 to E6, E6 to E7, E7 to D7, D7 to D8, and the destination. Arrive at.

  In the similarity method, a feature vector having pairs of adjacent meshes having directions as elements is generated for each destination based on the movement history. This feature vector corresponds to the prediction model in this method.

Then, in the prediction process, the similarity between the feature vector indicating the route from the departure place to the current position and the feature vector according to the prediction model is calculated, and it is determined which destination is likely. Specifically, the similarity between the feature vector indicating the route from the departure place to the current place and the feature vector according to the prediction model is calculated to determine which destination is likely. Note that the similarity of the feature vectors can be evaluated by an existing determination method, such as determining that the smaller the angle formed by the two vectors is, the more similar they are. Alternatively, the modified Jaccard coefficient may be used to calculate a similarity score, and a destination with a high similarity score may be determined to be likely. The modified Jaccard coefficient can be calculated by the following equation 3.

(4) Bayesian update method In the Bayesian update method, the arrival probability (also referred to as a likelihood function) to each destination (that is, a destination) with a condition that a certain mesh moves to an adjacent mesh is calculated, A probability table at the time of movement is generated for bidirectional movement between adjacent meshes. Then, for the destination estimated from the conditional probability table at the time of departure, the estimation correction based on the probability table at the time of movement is repeated. The conditional probability table at the time of departure is the same as that generated in the above (2) Bayesian network method. The conditional probability table at the time of movement is assumed to be the conditional probability table at the time of departure generated not for each departure place but for each traveled section.

  FIG. 14 is a diagram showing an example of a conditional probability table when moving. The conditional probability table at the time of movement represents the destination arrival probability for each section, day of the week, and time zone represented by a pair of adjacent meshes having a direction. FIG. 14 is a probability table related to the section from G2 to G3. When the movement history of the route shown in FIG. 14 is stored in the storage unit 11, in addition to the section from G2 to G3, G3 to F3, F3 to F4, F4 to F5, F5 to E5, E5 From E6 to E6, from E6 to E7, from E7 to D7, and from D7 to D8 in each probability table (that is, all transit intervals) The weighted visit count u (for example, Formula 2) is added. As described above, the conditional probability table at the time of departure and the conditional probability table at the time of movement (that is, the prediction model) are generated.

そして、エントロピーの最小値を更新した区間を特定し、特定された区間に係る曜日及び時間帯のセルを特異的区間とする。予測処理においては、このような特異的区間であるか否かという情報に基づいて処理を行うようにしてもよい。 In addition, in the present embodiment, when a certain section is passed, a section may be obtained such that the tendency to reach a specific destination becomes higher than a predetermined threshold value. Such a section is called a specific section. The specific section is obtained, for example, by calculating the entropy of each section in order from the place of departure for each record of the movement history. The entropy can be calculated by the following equation 4.

Then, the section in which the minimum value of the entropy is updated is specified, and the cells of the day of the week and the time zone related to the specified section are set as the specific section. In the prediction process, the process may be performed based on the information as to whether or not it is such a specific section.

  On the other hand, when predicting a destination while moving, the probability of reaching the destination obtained by the prediction process at the time of departure is used as the prior probability, and the prior probability is multiplied by the probability of reaching the destination for the new section. The posterior probability is calculated by (that is, Bayesian update). Bayes update is executed, for example, every time an adjacent mesh is entered. That is, the a priori probability and the posterior probability are relative, and the posterior probability calculated in a certain section is used as the a priori probability in the next section, and the posterior probability obtained by modifying the a priori probability is calculated. In this way, the destination arrival probability related to the present movement is corrected based on the past movement of the same section and the final destination while the moving body is moving. More specifically, the probability distribution indicating the a priori probability is multiplied by the likelihood function and further normalized to obtain the probability distribution indicating the posterior probability.

ただし、初回の移動時予測処理（Ｓ１４）では、出発時予測処理の結果を事前確率に用いるため、ベイズ更新は次の数６によって表される。

なお、Ｈ_iは、図示していない目的地ｉに向かうという仮説を示し、Ｐ（Ｈ_i）は出発時予測の結果として得られた目的地ｉに向かう確率である。また、Ｄ_BCは、メッシュＢを通過後、メッシュＣへ移動するという事象を示し、Ｐ（Ｄ_BC）は、メッシュＢを通過後、メッシュＣに移動する確率を示すものとする。すなわち、図９に示した移動履歴において、
通過点にメッシュＢからメッシュＣの順に移動したレコードの件数を、通過点にＢを含むレコードの件数で除した割合である。ここで、割合の算出に用いるレコードの件数（区間を移動した回数）は、例えば上述の数１に示したような重みづけした値ｕを用いるようにしてもよい。すなわち、履歴情報のうち時間的に新しい移動ほど影響が大きくなるように重みづけした疑似的な移動回数を用いて、ベイズ更新に係る処理を行うようにしてもよい。また、Ｐ（Ｄ_BC｜Ｈ_i）は、目的地ｉに向かっている場合において、メッシュＢを通過
した後、メッシュＣに移動する確率を示している。すなわち、図９に示した移動履歴において、到着情報のメッシュ又はＰＯＩが目的地ｉを示しているレコードのうち、通過点においてメッシュＢからメッシュＣに移動したレコードの件数の割合である。例えば、Ｐ（Ｄ_AB｜Ｈ_i）は、次の数７で表すことができる。

なお、Ｃ_ABは、図９の移動履歴において、メッシュＡの次にメッシュＢへ移動したレコード数をカウントした値である。数６では、Ｃ_ABからＣ_AIまでの８方向へ移動したレコードの数を用いている。また、数４のＰ（Ｈ_i｜Ｄ_AB）は、メッシュＡからメッシュＢへ移
動した場合において、目的地ｉに向かっている確率を示すものとする。すなわち、図９の通過点において、メッシュＡからメッシュＢへ移動したレコードのうち、到着情報のメッシュ又はＰＯＩが目的地ｉを示しているレコードの割合である。このような計算を各目的地について実行し、確率分布を更新する。 For example, the Bayesian update can be represented by the following equation 5. In addition, here, description will be given using meshes A to I shown in FIG.

However, since the result of the departure prediction process is used as the prior probability in the first travel time prediction process (S14), the Bayes update is expressed by the following formula 6.

It should be noted that H _i represents a hypothesis that the user is heading for a destination i (not shown), and P (H _i ) is a probability of heading for the destination i obtained as a result of the departure prediction. Further, it is assumed that D _BC indicates an event of moving to the mesh C after passing through the mesh B, and P (D _BC ) indicates a probability of moving to the mesh C after passing through the mesh B. That is, in the movement history shown in FIG.
It is a ratio obtained by dividing the number of records moved from the mesh B to the mesh C to the passing point in the order by the number of records including B at the passing point. Here, as the number of records used for calculating the ratio (the number of times the section is moved), for example, the weighted value u as shown in the above Expression 1 may be used. That is, the process related to the Bayesian update may be performed by using the pseudo number of movements weighted so that the more recent movements have a greater influence in the history information. Further, P (D _BC | H _i ) indicates the probability of moving to the mesh C after passing through the mesh B when heading for the destination i. That is, in the movement history shown in FIG. 9, it is the ratio of the number of records moved from the mesh B to the mesh C at the passing point among the records in which the mesh or POI of the arrival information indicates the destination i. For example, P (D _AB | H _i ) can be expressed by the following Equation 7.

Note that _CAB is a value obtained by counting the number of records moved to the mesh B after the mesh A in the movement history of FIG. 9. In Equation 6, the number of records moved in eight directions from _CAB to _CAI is used. In addition, P (H _i | D _AB ) in Equation 4 indicates the probability of moving toward the destination i when moving from the mesh A to the mesh B. That is, it is the ratio of the records in which the mesh or POI of the arrival information indicates the destination i among the records moved from the mesh A to the mesh B at the passing points in FIG. Such a calculation is executed for each destination to update the probability distribution.

When entering a section that has not been passed in the past, Bayes update may be performed using a preset default destination arrival probability instead of the prediction model. FIG. 16 is an example of a default destination arrival probability. The default destination arrival probability is, for example, a probability having the following property.
(1) As destinations, all destinations and “unknown” included in the movement history of the target moving body are included.
(2) Probabilities are given to all destinations and “unknown” included in the movement history of the target moving body, and the sum of these is set to 1.
(3) The probabilities for all destinations included in the movement history of the target moving body have the same value.
(4) The probability of “others” is twice the probability of all destinations included in the movement history of the target moving body.

  Further, it is determined whether or not the newly entered section is the above-mentioned specific section, and when it is determined that the newly entered section is the specific section, the destination arrival probability regarding the newly entered section may be multiplied. Good. On the other hand, when it is determined that it is not a specific section, Bayesian update may not be performed and the a priori probability may be directly used as the posterior probability.

  By storing the home and the destination estimated by the home estimation process and the destination estimation process in the storage unit 11, the estimated destination information and the estimated home position are stored in S11 and S12 of FIG. You will be able to get it. By using the information estimated from the movement history, the user does not need to input troublesome information.

<Other>
The present invention includes a computer program that executes the above processing. Further, a computer-readable recording medium recording the program also belongs to the category of the present invention. With respect to the recording medium in which the program is recorded, the above-described processing becomes possible by causing a computer to read and execute the program in the recording medium.

  Here, the computer-readable recording medium refers to a recording medium that can store information such as data and programs by an electrical, magnetic, optical, mechanical, or chemical action and can be read by a computer. Among such recording media, those removable from the computer include flexible disks, magneto-optical disks, optical disks, magnetic tapes, memory cards and the like. In addition, a hard disk drive, a ROM, or the like is used as the recording medium fixed to the computer.

  In addition, the habituation determination may not be executed when the number of arrivals that is equal to or more than a predetermined number is not held for the corresponding time zone of the corresponding day of the week. This is because the accuracy of the judgment cannot be obtained unless past results have been accumulated to some extent.

  As shown in FIG. 1, the moving body is not limited to a vehicle such as a passenger car, and may be a user having a smartphone or the like. In this case, the moving means may be estimated based on the moving speed or the position information, and information corresponding to the moving means may be output. For example, it is possible to output appropriate information so as to propose different sightseeing plans for traveling by vehicle and traveling by train.

DESCRIPTION OF REFERENCE NUMERALS 1 movement status determination device 11 storage unit 12 position information acquisition unit 13 movement history generation unit 14 model generation unit 15 destination acquisition unit 16 home position acquisition unit 17 out-of-life determination unit 18 habit determination unit 19 information output unit 2 network 3 (3a, 3b ...) Moving body

Claims (6)

A movement history generation unit that generates a movement history including a visited location of the mobile body based on position information indicating the position of the mobile body,
Based on the movement history, the model generation unit that aggregates the number of times of arrival at each visited destination related to the past movement of the mobile body for each arrival time zone,
The destination of the mobile body, which is a location estimated based on the travel history and the travel route from the departure point of the mobile body to the current location, has the largest number of past arrivals in the estimated arrival time zone. If it is a visited place, a habituation determination unit that determines that the movement of the mobile body is a habitual action,
An information output unit that outputs information to the mobile body according to the determination of the habit determination unit,
A traveling condition determination device having. When the starting point and the destination of the moving body included in the moving history are within a predetermined range from the position of the user's home related to the moving body, or the starting point and the destination of the moving body However, when the passage history is included in a section equal to or more than a threshold in the movement history, the vehicle further includes an outside-living-area determining unit that determines that the movement is within the living area,
The movement status determination device according to claim 1, wherein the information output unit outputs information to the moving body according to the determination of the habit determination unit and the determination of the outside-of-living-range determination unit. The movement status determination device according to claim 2 , wherein the position of the user's home related to the moving body is a place estimated based on the movement history. The movement history according to any one of claims 1 to 3 , wherein the movement history represents a transition of position information of the moving body based on a mesh that divides an area into a mesh shape having a predetermined position and size. Judgment device. A movement history generating step of generating a movement history including a destination of the mobile body based on position information indicating the position of the mobile body;
Based on the movement history, the model generation step of totaling the number of times of arrival at each destination related to the past movement of the mobile body for each arrival time zone,
The destination of the mobile body, which is a location estimated based on the travel history and the travel route from the departure point of the mobile body to the current location, has the largest number of past arrivals in the estimated arrival time zone. If it is a visited place, the habituation determination step of determining that the movement of the mobile body is a habitual action,
An information output step of outputting information to the mobile body according to the determination made in the habit determination step ,
A method for determining a movement status that a computer executes. A movement history generating step of generating a movement history including a destination of the mobile body based on position information indicating the position of the mobile body;
Based on the movement history, the model generation step of totaling the number of times of arrival at each destination related to the past movement of the mobile body for each arrival time zone,
The destination of the mobile body, which is a location estimated based on the travel history and the travel route from the departure point of the mobile body to the current location, has the largest number of past arrivals in the estimated arrival time zone. If it is a visited place, the habituation determination step of determining that the movement of the mobile body is a habitual action,
An information output step of outputting information to the mobile body according to the determination made in the habit determination step ,
A program that causes a computer to execute.

Images