JP6136992B2 - Action area prediction device - Google Patents

Description

  The present invention relates to a behavior area prediction apparatus that predicts a behavior area such as a moving object.

  In general, in a route information processing system mounted on a vehicle, a so-called navigation system, route guidance from a current location to a destination location is performed by voice or image. In route guidance by such a navigation system, first, a destination is set by a driver or the like via an input device of the navigation system. On the other hand, in driving assistance by the navigation system, a route from the current point of the vehicle to the set destination point is guided to the driver and the like along with surrounding traffic information.

  By the way, a driver often does not bother to set a destination point for a navigation system, such as a destination point to reach through a familiar route such as a commute or home. Even without this, a technique for predicting the destination point has been proposed. An example thereof is described in Patent Document 1.

  The technique described in Patent Document 1, that is, the information processing apparatus described in the document, is an action that generates action prediction information including a departure place, a destination, a route from the departure place to the destination, and the like based on the action history. A prediction server; and a behavior prediction information display control unit that acquires the generated behavior prediction information and displays the information on the display unit together with the map information. Then, the behavior prediction information display control unit causes the display unit to display the behavior prediction information so as to be within the displayable range of the display unit with the change of the acquired behavior prediction information.

JP 2011-214948 A

  In the information processing apparatus described in Patent Document 1, a destination point is predicted by selecting one or a plurality of destination points from an action history including the destination point. In recent years, it has become necessary to predict such destinations in an easy-to-understand manner. However, when selecting from past destination points, the predicted destination points are also subject to unnecessary restrictions due to past action history. Not necessarily.

  In addition, in the mobile body which is not restricted to a vehicle and the path | route is not prescribed | regulated, the above-mentioned subject concerning the prediction of the destination point is also a thing common in general.

  The present invention has been made in view of such circumstances, and its purpose is to predict the action area more appropriately while using the past action history and suppressing the restrictions due to the past action history. It is in providing the action area prediction apparatus which can do.

  Hereinafter, means for solving the above-described problems and the effects thereof will be described.

  In order to achieve the above object, the behavior area prediction device is a behavior area prediction device that predicts a behavior area including an area that is a destination point based on past behavior history information associated with map information. A divided area placement unit for placing a divided area, which is an area of a predetermined size, on a map of map information corresponding to the action history information of the past, and past destination points included in the divided area for each of the divided areas A probability calculating unit that calculates a probability that the divided area is an area including the new destination point based on the calculated probability, and a probability calculating unit that calculates the probability that the divided area is an area including the new destination point, and the plurality of divided areas The area of the action area is large enough to include one or more segment areas selected so that the probability of becoming an area including a new destination point is equal to or greater than a predetermined value. And summarized in that and a specifying unit for specifying a of.

  According to such a configuration, the size of the area of the action area is specified based on a plurality of divided areas selected so that the probability of becoming an area including a new destination point is equal to or greater than a predetermined value. Thereby, it becomes possible to predict the action area not including the past destination point but including the area including the new destination point. Therefore, it becomes possible to predict the action area more appropriately while using the past action history (the destination point) and suppressing the restrictions on the past action history (the destination point). For example, when going to a meal, even if the destination point is not specified, an action area whose size is specified so as to include an area including a point where it is customary to go to a meal will be predicted. . That is, even when the destination point is not clearly defined, it is possible to specify an action area including an area having characteristics expected by the driver.

  Preferably, the specifying unit selects the action areas in order from a segment area having a high probability of becoming an area including a new destination point.

  According to such a configuration, the identified action area can be narrowed down to an area including a partitioned area with a high probability of becoming an area including a new destination point. As a result, the probability of becoming an area including a new destination point in the action area can be increased, and the section area selected as the action area can be narrowed down.

  Preferably, the divided area arrangement unit sets the divided area as a circular area, and includes a center point of each divided area arranged at equal intervals on the map when the divided area is arranged. The divided areas are rearranged at positions including the most past destination points.

  According to such a configuration, since more past destination points can be included in the divided area, the possibility that the divided area can be partitioned as an area having characteristics expected by the driver is increased.

  Preferably, the information on the action history is associated with information indicating the number of times the user has visited the destination in the past, and the specifying unit includes the destination point included in the section area for the section area included in the action area. And the size of the divided area is reset based on the distance from the center to the destination point where the accumulated number of stops reaches the stop count threshold.

  According to such a configuration, the divided areas included in the action area are reset so that the area is narrowed according to the number of times of stopping at the past destination points. For example, if the segmented area is circular, the radius is reset based on the distance from the center of the segmented area to the destination point where the number of stops reaches the frequency threshold so that the area of the segmented area becomes smaller. Is set. Also, if the segmented area is a rectangle, the length of each side is the ratio of the distance from the center of the segmented area to the destination point where the number of stops reaches the number of times threshold and the distance to the side on the extension. Based on the resetting, the area of the divided area is reset to be smaller. That is, the area of the area is narrowed down (the size is reduced) while maintaining a high probability that the action area becomes an area including a new destination point.

  Whether or not the accumulated number of visits has reached the threshold for the number of visits may be determined by, for example, comparison with the number of times threshold set as a predetermined number of times. You may judge by comparing with the ratio with respect to the frequency | count of the stop of the past destination points more than predetermined number, such as all contained in the area.

  In addition, since the area of the action area can be reduced (smaller in size), information related to driving support can be provided more appropriately. For example, if the area of the action area is small, the scale can be increased in the map display, so that more information can be provided.

  Preferably, the action area prediction device displays a display device capable of displaying an image of the map information and the action area together, and the display device displays a predetermined number or more of divided areas included in the action area. And a display control unit for controlling the display scale of the map information.

  According to such a configuration, since a predetermined number or more such as all of the predicted action areas are displayed in the display area, it is preferable to provide information related to driving support related to the predicted action area to the driver. Will be able to. In addition, when the action area is displayed on the display device in this way, it is possible to provide information on a partitioned area with a high probability of becoming an area including a new destination point, that is, information on the action area at an earlier stage.

  Preferably, the display control unit further controls the display scale so that the current location is also displayed.

  According to such a configuration, it is possible to overlook an area with a high probability of being a route, and it is also possible to provide information about such an area, so that reduction of the scale changing work by the driver is expected. This will improve the convenience for the driver.

  Preferably, the display control unit moves the map range to the action area while displaying a predetermined number or more of the divided areas included in the action area on the display device under the controlled display scale. The display range is set and controlled in such a manner that an area that has a high probability of including a new destination point is displayed together even if it is a divided area that is not included.

  According to such a configuration, in the display range, together with the predicted action area, an area that has a high probability of becoming a new destination point out of areas not included in the action area is displayed together. The probability of including a new destination point in the display range is increased.

  Preferably, the display control unit further controls the display range in such a manner that the most past destination points not included in the action area are included.

  According to such a configuration, the probability that the display range includes a new destination point in the display range is increased by including more past destination points together with the predicted action area and the like.

  Preferably, the probability calculation unit calculates a probability that a past destination point becomes a new destination point using a Bayesian network.

  According to such a configuration, the probability of becoming a new destination point can be suitably calculated based on various events. In addition, since the degree of freedom for an event that can be used as a cause increases, it is expected that the probability of becoming a new destination point can be appropriately calculated.

  Preferably, the probability calculation unit uses the Bayesian network so as to include at least one event among time, day of the week, and starting point as an event that causes “Basis” of the Bayesian network.

  According to such a configuration, it is expected that the calculation accuracy of the probability of becoming a new destination point can be improved by using a parameter for an event highly relevant to movement.

Schematic which shows the schematic structure of the system about one Embodiment which actualized the action area prediction apparatus. The flowchart which shows the outline | summary about the flow of the process which estimates the action area in the embodiment. The flowchart which shows the flow of the process regarding arrangement | positioning of the division area in the embodiment. The schematic diagram which shows typically the process regarding arrangement | positioning of the division area in the embodiment. The flowchart which shows the flow of the process which calculates the probability which becomes an area containing the new destination point of the division area in the embodiment. The schematic diagram which shows typically the process which calculates the probability which becomes an area containing the new destination point of the division area in the embodiment. The flowchart which shows the flow of the process which estimates the action area in the embodiment. The schematic diagram which shows typically the process which estimates the action area in the embodiment. The flowchart which shows the flow of the process which makes the area | region of the action area small in the embodiment. The schematic diagram which shows typically the process which makes the area | region of the action area small in the embodiment. The flowchart which shows the flow of the process which calculates the display scale of the action area in the embodiment. The schematic diagram which shows typically the process which calculates the display scale of the action area in the embodiment. The flowchart which shows the flow of the process which controls the display position of the action area on the display screen in the embodiment. The schematic diagram which shows typically the flow of the process which controls the display position of the action area on the display screen in the embodiment.

  One embodiment of the action area prediction apparatus will be described with reference to FIGS. First, a schematic configuration of the action area prediction apparatus will be described with reference to FIG.

  As shown in FIG. 1, the action area prediction apparatus includes a vehicle 10 as a moving body and an information center 30 that provides various types of information to the vehicle.

  The information center 30 includes a device such as a computer that can execute various types of information processing, a communication device that enables information communication with the outside, and the like. That is, the information center 30 collects information to be provided to the vehicle 10 and the center communication device 31 that exchanges various information with the vehicle 10 via wireless communication or the like, and communicates the collected information to the vehicle 10. And an information providing unit 32 that performs the process of providing via the information. The information center 30 includes a nonvolatile storage device. Various information acquired by the information providing unit 32 is stored in the storage device. Specifically, the storage device includes a traffic information DB (database) 33 including regulation information and traffic jam information on the road, a weather information DB 34 including information such as temperature, precipitation, and sunshine in the area, road information, A map information DB 35 including information related thereto is provided. The traffic information DB 33 holds information appropriately acquired by the traffic information center, the weather information DB 34 holds information appropriately acquired from the weather information center, and the map information DB 35 is stored in advance, Information that is appropriately updated by the information center is held. That is, the information center 30 is configured to be able to communicate with a traffic information center, a weather information center, a map information center, and the like.

  The vehicle 10 is a moving body such as an automobile that moves by a driving operation of the driver.

  The vehicle 10 includes a driving support ECU (Electronic Control Unit) 11 that generates and outputs driving support information. In addition, the vehicle 10 provides a navigation device 20 that provides a driving route guidance to the driver and the like to the driver, a display device 21 that displays driving support information and the like in a manner that is visible to the driver, and the like. And a speaker 22 for outputting driving support information and the like in an audible manner. Further, the vehicle 10 includes an in-vehicle communication device 27 that exchanges various information with the information center 30 via wireless communication or the like, a GPS (global positioning system) device 28 for specifying the current position of the vehicle 10, a vehicle And a vehicle speed sensor 29 that measures ten travel speeds. The driving assistance ECU 11 is connected to the navigation device 20, the in-vehicle communication device 27, the display device 21, the speaker 22, the GPS device 28, and the vehicle speed sensor 29 through an in-vehicle network such as a CAN (controller area network). Therefore, the navigation device 20 and the in-vehicle communication device 27 are connected to the driving support ECU 11 in a manner that allows various information to be exchanged. The driving support ECU 11 is connected to a display device 21 and a speaker 22 in a manner capable of transmitting various types of information. Furthermore, a GPS device 28 and a vehicle speed sensor 29 are connected to the driving support ECU 11 in a manner capable of receiving various types of information. In addition, although not shown in figure, the required information communication is possible also between each apparatuses other than driving assistance ECU11.

  The navigation device 20 refers to map information including road information stored in advance and guides a travel route to the destination by setting the current location of the vehicle 10 and the destination that is a specific location. The navigation device 20 specifies the current location by inputting the position information output from the GPS device 28. In addition, the navigation apparatus 20 may be provided with the GPS apparatus uniquely. Moreover, the navigation apparatus 20 is provided with the operation part, and a destination point is set by operation by the driver etc. via the said operation part. Note that the operation unit may be provided separately from the navigation device 20. In addition, the navigation device 20 can set a destination point based on information on the destination point input from the driving support ECU 11. Furthermore, in the present embodiment, the navigation device 20 can set an area having a predetermined area as a destination of the vehicle 10, that is, a so-called action area, in addition to the destination point. And when an action area is set, the navigation apparatus 20 can guide the moving route to the set action area. In addition, such an action area is set based on information input from the driving support ECU 11. Therefore, the navigation device 20 is configured to perform route guidance to the action area by setting the action area even when the destination point is not specified. That is, the navigation device 20 can provide the driver with various types of information such as traffic information and event information as well as travel route guidance for traveling to the action area.

  The navigation device 20 stores map information including road information, various travel models, and the like used for route guidance in a non-volatile storage device such as an HDD (hard disk drive). The map information including the road information may be shared with the driving support ECU 11. The map information including road information is information relating to the map, and is composed of data for displaying the background of the road and the road map, data including the name of the intersection, and the like. Further, the map information including the road information includes road ancillary information such as road shape, information on intersections and crosswalks on the road. Furthermore, the map information including road information includes information on various facilities together with information such as an address and latitude / longitude that can identify the location. Various facilities include offices such as houses, commercial facilities and factories, and public facilities such as government offices and parks.

  In addition, the navigation device 20 can appropriately output various information input by the operation unit, information on the current position of the vehicle 10 and information on the destination point to the driving support ECU 11. Therefore, the driving assistance ECU 11 can acquire the movement history of the vehicle 10.

  The display device 21 is configured by a liquid crystal display or the like, and is installed at a position where it can be viewed by the driver. The display device 21 displays information related to the guidance of the moving route as an image or the like. For example, the navigation device 20 generates image data that combines the current location of the vehicle 10 and a map around it, and causes the display device 21 to display the generated image data.

  The speaker 22 is a device that generates sound and sound, and outputs sound and sound corresponding to sound and sound data input from the navigation device 20. The navigation device 20 outputs voice information such as route guidance and traffic information through the voice device.

  The in-vehicle communication device 27 secures a wireless communication line with the information center 30 by wireless communication such as a public line or a dedicated line, and enables various information to be exchanged with the information center 30. The in-vehicle communication device 27 may be configured to communicate directly with the information center 30, or may be configured to communicate with the information center 30 through various networks or communication lines such as one or a plurality of internets in the route. .

  The GPS device 28 receives a signal from a GPS satellite and detects the current location of the vehicle 10 based on the received signal from the GPS satellite. The GPS device 28 outputs position information indicating the detected current location to the driving assistance ECU 11 and the navigation device 20.

  The vehicle speed sensor 29 detects the rotational speed of the wheels of the vehicle 10 and outputs a signal corresponding to the detected rotational speed to the driving assistance ECU 11 and the navigation device 20. For example, when the signal corresponding to the speed “0” is output from the vehicle speed sensor 29, it is specified that the vehicle 10 is stopped.

  The driving assistance ECU 11 includes a microcomputer having a calculation device (CPU) and a storage device. More specifically, the driving support ECU 11 temporarily stores an arithmetic unit that executes arithmetic processing of a control program, a read-only memory (ROM) that stores the control program and data, and an arithmetic result of the arithmetic unit. Volatile memory (RAM). The driving support ECU 11 includes a readable / writable nonvolatile storage device including an HDD (hard disk drive), a flash memory, and the like. As a result, the driving support ECU 11 reads the control program stored in the read-only memory or the storage device into the arithmetic device and executes it to perform the necessary information processing. In the present embodiment, the driving support ECU 11 holds a program for predicting an action area, and a process for predicting the action area is performed by executing the program on an arithmetic device or the like.

  As shown in FIG. 2, in the present embodiment, the driving support ECU 11 performs a segment area arrangement process (step S <b> 10) that arranges a segment area around the current location of the vehicle 10 as a process for predicting an action area, Probability estimation processing (step S20) is performed for estimating the probability of an area including a new destination point for the area. Further, the driving support ECU 11 performs a prediction process (step S30) for predicting an action area from among the divided areas, an action area narrowing process (step S40), a display scale calculation process (step S50), and a display. A display range setting process (step S60) for adjusting the display range including the action area displayed on the screen is performed. And driving assistance ECU11 will control the map display by the navigation apparatus 20 so that the display apparatus 21 may display the range containing an action area, if a display range is defined. Note that the driving assistance ECU 11 may control the display device 21 so that the range including the action area is displayed on the display device 21.

  Further, as shown in FIG. 1, the driving support ECU 11 determines a behavior area from among the divided areas, a divided area arranging unit 111 that performs divided area arrangement processing, a probability calculating unit 112 that performs probability estimation processing for each divided area, and the divided areas. And a prediction unit 113 as a specific unit for prediction. The driving support ECU 11 also includes a narrowing processing unit 114 that performs a narrowing process of the action area, a display scale calculation unit 115 that performs a display scale calculation process, and a display position of the action area displayed on the display screen. And a display range setting unit 116 that performs display position adjustment processing to be adjusted. Note that the display control unit is configured to include a display scale calculation unit 115 and a display range setting unit 116.

  The divided area arrangement unit 111 determines the periphery of the vehicle 10 within a predetermined distance from the current location of the vehicle 10 by the divided area arrangement process, and sets a plurality of areas on the map corresponding to the periphery of the determined vehicle 10. By dividing into areas, one or a plurality of divided areas are arranged on the map.

  The probability calculation unit 112 calculates the probability that each divided area becomes an area including a new destination point of the vehicle 10 for a plurality of divided areas by the probability estimation process. In the present embodiment, a probability that a past destination point included in the segment area becomes a new destination point is calculated using a Bayesian network, and the segment is determined based on the calculated probability of becoming a new destination point. The probability that the area is an area including a new destination point is calculated. The Bayesian network is a model that can indicate a causal relationship with probability, and by using the “cause” and “result” of the causal relationship as input values, the ratio at which the “result” is generated according to the “cause” is calculated. In addition, the probability calculation unit 112 uses “time”, “day of the week”, and “departure point” as the “cause” event, and “target point” as the “result” event as the input value of the Bayesian network. .

  The prediction unit 113 has a higher probability of becoming an area including a new destination point from among a plurality of divided areas, based on the probability of becoming an area including a new destination point calculated for each divided area by the prediction process. A division area is selected, and the selected division area is predicted as an action area. That is, the prediction unit 113 sets the size of the action area so as to include one or more segment areas selected so that the probability of becoming an area including a new destination point is greater than or equal to a predetermined value from the plurality of segment areas. Specify the size.

  The narrowing processing unit 114 reduces the area size while maintaining a high probability that the predicted action area will be an area including a new destination point by the narrowing processing, that is, the size of the area. Make it smaller. More specifically, each divided area is narrowed apart from the divided areas included in the action area. Then, the action areas are narrowed by rearranging the narrowed section areas on the map.

  The display scale calculation unit 115 calculates the display scale of the action area so that it can be displayed on the display device 21 by the display scale calculation process. That is, the driving assistance ECU 11 controls the display of the action area to the display scale.

  The display range setting unit 116 adjusts the display position of the action area displayed on the display screen by the display position adjustment process so that the probability that the entire display screen includes a new destination point is high. That is, the driving assistance ECU 11 controls the display range including the action area on the display device 21.

  Further, the driving support ECU 11 includes a movement history management unit 117 that manages history information including a past departure position of the vehicle 10 and a past destination point Tg. Furthermore, the driving assistance ECU 11 includes an action history DB (database) 121, a traffic history DB 122, and a map information DB 123 in its storage device. Therefore, when predicting the action area, the driving support ECU 11 can access each DB as necessary, and can write and read necessary information.

  The action history DB 121 stores the past departure point, destination point, route from the departure point to the destination point, and the like of the vehicle 10 together with time, day of the week, weather, traffic conditions, and other peripheral information as an action history. Information such as a departure point, a destination point, a route from the departure point to the destination point, and information such as time, day of the week, weather, traffic conditions, and other peripheral information are acquired from the navigation device 20. The navigation device 20 can acquire weather, traffic conditions, and other peripheral information from the information center 30.

  In the traffic history DB 122, traffic conditions around routes and points not included in the behavior history are stored together with information such as time, day of the week, and weather. These pieces of information are acquired from the information center 30 and stored in a timely manner.

  The map information DB 123 stores map information for route guidance and display. Information included in the map information is acquired from the information center 30 and updated with new information as appropriate.

  The action history DB 121, the traffic history DB 122, and the map information DB 123 may be shared with the navigation device 20.

  The movement history management unit 117 can perform processing for storing the past behavior history of the vehicle 10 in the behavior history DB 121. Moreover, the movement history management unit 117 can perform a process of reading a past action history required for the process of predicting the action area from the action history DB 121. The movement history management unit 117 displays the departure point, destination point, route information from the departure point to the destination point acquired from the navigation device 20, various traffic information such as speed in the route, time, day of the week, weather, etc. Store it in the DB 122. The movement history management unit 117 deletes the past action history that has already been stored from the old one, or performs integration processing so that a new past action history is retained in the traffic history DB 122. To do.

  In addition, when one or more search conditions such as a departure point, a destination point, a time, a day of the week, and weather are input, the movement history management unit 117 moves in the past corresponding to the input one or more search conditions. The history can be extracted from the traffic history DB 122 and acquired.

  Next, a process for predicting an action area will be described with reference to FIGS. In the following, for convenience of illustration, FIGS. 4, 6, 8, 10, 12, and 14 show only the area corresponding to the display area 211 of the display device 21 among the surrounding areas of the vehicle 10, and the explanation thereof will be made. For convenience, processing for predicting an action area only for the indicated area will be described. 4, 6, 8, 10, 12, and 14, the past destination point Tg is indicated by a rectangle and the current point PC of the vehicle 10 is indicated by a filled triangle.

  First, with reference to FIGS. 3 and 4, the divided area arrangement process (step S <b> 10 in FIG. 2) will be described. When the divided area arrangement process is started, the driving assistance ECU 11 temporarily arranges divided areas having a predetermined size in the surrounding area of the vehicle 10 at regular intervals (step S11 in FIG. 3). For example, as shown in FIG. 4, the driving assistance ECU 11 arranges temporary center points P11 to P15 at equal intervals in the surrounding area of the vehicle 10, and has a circular shape with a predetermined radius r1 around each temporary center point P11 to P15. Temporarily arrange the area as a partitioned area. For example, the surrounding area of the vehicle 10 is a range with a radius of 10 km centered on the current point PC, and the temporary center points P11 to P15 are arranged at intervals of 4 km in the east, west, north, and south directions (up, down, left and right in the figure). The area has a predetermined radius r1 of 2 km.

  The driving assistance ECU 11 selects one of a plurality of temporarily arranged divided areas (step S12 in FIG. 3). Note that once the selected area is selected, it is not selected again. And driving assistance ECU11 moves the arrangement | positioning position of the division area temporarily arrange | positioned in the range containing the corresponding temporary center point about the selected division area, and the center point containing the most past destination points Tg is shown. Search is performed (step S13 in FIG. 3). Further, the driving assistance ECU 11 determines the center point searched as including the most past destination points Tg as the position where the divided area is arranged (step S14 in FIG. 3). As a result, the divided areas whose center points are determined are arranged. Then, the driving assistance ECU 11 determines whether or not, for example, all the divided areas have been selected as a predetermined number or more (step S15). If it is determined that there is an unselected segment area (NO in step S15 in FIG. 3), the driving assistance ECU 11 returns the process to step S12, selects the next segment area, and selects the selected segment area. Processing for determining the arrangement is performed (steps S12 to S14 in FIG. 3). On the other hand, for example, if it is determined that all the divided areas have been selected (YES in step S15 in FIG. 3), the first to fifth corresponding to all the first to fifth divided areas C11 to C15 temporarily arranged. Since the arrangement of the divided areas C21 to C25 is confirmed, the driving assistance ECU 11 ends the divided area arrangement process.

  That is, as shown in FIG. 4, for example, the driving assistance ECU 11 selects one fourth division area C14 from the temporarily arranged first to fifth division areas C11 to C15. In FIG. 4, for convenience of illustration, only the fourth section area C14 temporarily positioned corresponding to the temporary center point P14 is illustrated, and temporarily arranged corresponding to the other temporary center points P11 to P13, P15. Illustration of each divided area consisting of a circular area is omitted. When the driving assistance ECU 11 moves the arrangement position of the fourth divided area C14 within the range including the temporary center point P14 for the selected fourth divided area C14, the most past destination points Tg. It is searched that the center point at which is included is the center point P24. Therefore, the driving assistance ECU 11 determines the searched center point P24 as a position where the fourth division area C24 corresponding to the fourth division area C14 is arranged. Thereby, the fourth divided area C24 in which the center point P24 is determined is arranged. Similarly, corresponding to the first segment area C11, the first segment area C21 including the temporary center point P11 and centering on the center point P21 having the largest number of past destination points Tg is arranged. Corresponding to the second section area C12, a second section area C22 including the temporary center point P12 and centering on the center point P22 having the largest past destination point Tg is arranged. Corresponding to the third segment area C13, a third segment area C23 including the temporary center point P13 and centering on the center point P23 having the largest past destination point Tg is arranged. Corresponding to the divided area C15, a fifth divided area C25 including the temporary center point P15 and centering on the center point P25 having the largest number of past destination points Tg is arranged.

  Next, with reference to FIG. 5 and FIG. 6, the probability estimation processing (step S <b> 20 in FIG. 2) of each divided area will be described. Since the probability estimation process may be performed separately for each divided area, the case where the probability estimation process is performed for one divided area will be described below, and the description of the probability estimation process for the other divided areas will be omitted. In the present embodiment, processing using a Bayesian network is performed as probability estimation processing.

  When the probability estimation process is started, the driving support ECU 11 extracts all the past destination points Tg included in the one divided area and extracts the extracted past destination points. Information associated with each Tg is set as an input value for probability estimation processing (step S21 in FIG. 5). Information related to the past destination point Tg is acquired from the action history DB 121 via the movement history management unit 117. Further, in the present embodiment, since the Bayesian network is used for the probability calculation process, based on the fact that the “cause” element and the “result” element are set as a set of input values, The probability of occurrence of “result” from “cause” is calculated. That is, such a set of input values is set for each past destination point Tg for each past destination point Tg. The “cause” element includes “time”, “day of the week”, “departure point”, and the “result” element includes “target point”. Then, the driving assistance ECU 11 performs a Bayesian network process based on the input value (step S22 in FIG. 5). For example, from the state of each element of “time”, “day of the week”, and “departure point”, the probability that a “destination point” occurs, that is, the probability that it becomes a new destination point is calculated. For each past destination point Tg by Bayesian network processing, one probability that the “destination point” becomes a new destination point is calculated. Then, the driving assistance ECU 11 adds all the calculated probabilities of becoming a new destination point, and uses the added value as an estimated value for the probability PR of becoming an area including the new destination point corresponding to the one divided area. Confirm (step S23 in FIG. 5). Then, the probability estimation process for one target divided area is ended. Note that the driving support ECU 11 executes probability estimation processing for all the divided areas, and determines an estimated value of the probability PR that becomes an area including a new destination point for all the divided areas.

  As shown in FIG. 6, for example, by such probability estimation processing, the probability PR that becomes an area including a new destination point corresponding to the first divided area C21 is 51%, and the new probability corresponding to the second divided area C22 is new. The probability PR of becoming an area including the destination point is determined as 4%, and the probability PR of being an area including a new destination point corresponding to the third divided area C23 is determined as 3%. In addition, the probability PR of an area including a new destination point corresponding to the fourth division area C24 is 32%, and the probability PR of an area including a new destination point corresponding to the fifth division area C25 is 7%. Confirmed. In this case, the first segment area C21, the fourth segment area C24, the fifth segment area C25, the second segment area C22, the third segment in descending order of the probability PR of becoming an area including a new destination point. It becomes area C23.

  Next, with reference to FIG.7 and FIG.8, the prediction process (step S30 of FIG. 2) of an action area is demonstrated. The action area prediction process is a process that is performed by comparing, for example, all the divided areas with a predetermined number or more, and is performed once for all the divided areas.

  When the action area prediction process is started, the driving assistance ECU 11 arranges the probabilities PR to be areas including new destination points in the first to fifth divided areas C21 to C25 in descending order of the probability (step of FIG. 7). S31). Further, the driving support ECU 11 accumulates the cumulative value obtained by accumulating the probability PR that becomes an area including the new destination point in descending order until the cumulative value becomes equal to or higher than the probability threshold value as a predetermined value (step S32 in FIG. 7). Then, the section area corresponding to the probability PR that becomes the area including the new destination point added to the cumulative value that is equal to or higher than the probability threshold is selected as the action area (step S33 in FIG. 7). Thereby, an action area including the selected divided area as an area serving as a destination point is predicted.

  That is, as shown in FIG. 8, the driving assistance ECU 11 places the first division area C21 in the first place, the fourth division area C24 in the second place, and the second place in the descending order of the probability PR of becoming an area including a new destination point. The fifth section area C25 is third, the second section area C22 is fourth, and the third section area C23 is fifth. And driving assistance ECU11 arranges the probability PR which becomes an area including the new destination point of the 1st-5th division areas C21-C25 in this order. That is, the probabilities PR of areas including new destination points are arranged in the order of “51%”, “32%”, “7%”, “4%”, and “3%”. Then, the cumulative value of the probability PR that becomes an area including the new destination point accumulated in this order is compared with the set probability threshold value. For example, when the probability threshold is set to “80%”, “51%” and “32%” are accumulated to be “80%” or more. Based on this result, the first divided areas corresponding to the probabilities PR “51%” and “32%” that are areas including each new destination point accumulated until the probability threshold is “80%” or more. An action area including C21 and the fourth divided area C24 as an area serving as a destination point is predicted.

  Next, the action area narrowing process (step S40 in FIG. 2) will be described with reference to FIG. 9 and FIG.

  When the action area narrowing process is started, the driving support ECU 11 selects one of the divided areas included in the action area (step S41 in FIG. 9). At this time, the once selected section area is not selected again. In addition, the total number of visits associated with all past destination points Tg included in the selected segment area is calculated as the total number of visits, and the accumulated visits that hold the accumulated value are reset to zero. Then, for the selected divided area, one past destination point Tg close to the center point is selected (step S42 in FIG. 9). The past destination point Tg once selected is not selected again. Next, the past number of stops associated with the selected past destination point Tg is accumulated in the accumulated number of stops (step S43 in FIG. 9). Then, the driving support ECU 11 determines whether or not the “cumulative number of stops / total number of stops” has reached the number threshold or more, that is, has reached the number threshold (step S44 in FIG. 9). If it is determined that “cumulative number of stops / total number of stops” is not equal to or greater than the number threshold (NO in step S44 in FIG. 9), the driving support ECU 11 returns the process to step S42 and stops at the next past destination point Tg. A process of accumulating the number of times into the cumulative number of visits is performed. On the other hand, when it is determined that “cumulative number of stops / total number of stops” is equal to or greater than the number threshold (YES in step S44 in FIG. 9), the driving support ECU 11 determines the past purpose in which the number of stops was added last from the center point. The distance to the point Tg is calculated (step S45 in FIG. 9). Further, the driving assistance ECU 11 sets a narrow area where the section area is narrowed so that the calculated distance is a radius (step S46 in FIG. 9). In other words, the narrow area is an area obtained by resetting the radius of the divided area to a radius having a smaller calculated distance.

  Then, it is determined whether or not a predetermined number or more, for example, all divided areas included in the action area have been selected (step S47 in FIG. 9). If it is determined that not all the divided areas included in the action area are selected (NO in step S47 in FIG. 9), the driving assistance ECU 11 returns the process to step S41, selects the next divided area, and selects the selected area. Set a narrow area in the divided area. On the other hand, if it is determined that all the divided areas included in the action area have been selected (YES in step S47 in FIG. 9), the size of each divided area included in the action area is reset to the size of the narrow area. Then, the range of the action area is adjusted to be a narrower range based on each divided area having the size of the narrow area (step S48 in FIG. 9). As a result, the range of the action area is also narrowed, and the action area narrowing process ends.

  That is, as shown in FIG. 10, when the driving assistance ECU 11 selects the first segment area C21, first, the driving assistance ECU 11 first calculates the total number of stops by accumulating the number of stops associated with all past destination points Tg. . For example, “90%” is set as the number-of-times threshold. Then, the number of visits associated with the past destination point Tg close to the center point P21 is sequentially accumulated to obtain the cumulative number of visits. Specify TGx. Then, the narrow area C31 having the radius rx as the distance from the center point P21 to the past destination point TGx is specified. That is, the radius rx of the narrow area C31 is set to a length shorter than the initial radius r1. Similarly, when selecting the fourth segment area C24, the driving assistance ECU 11 sequentially accumulates the number of stoppages associated with the past destination point Tg close to the center point P24 to obtain the cumulative number of stoppages. A past destination point TGy for which the “total number of stops” is equal to or greater than the threshold is specified. Then, the narrow area C34 having the radius ry as the distance from the center point P24 to the past destination point TGy is specified. That is, the radius ry of the narrow area C34 is set to a length shorter than the initial radius r1. The range of the action area is also set to the initial size by the first divided area C21 changed to the size of the narrow area C31 and the fourth divided area C24 changed to the size of the narrow area C34. In comparison, it becomes narrower.

  Next, display scale calculation processing (step S50 in FIG. 2) will be described with reference to FIGS. The display scale calculation process is performed on the entire action area. In this way, by evaluating the past destination points Tg based on the number of stops, the past destination points Tg are processed so as to increase the weight of the points that have a high probability of becoming new destination points. The narrowing of the area is appropriately performed.

  When the display scale calculation process is started, the driving support ECU 11 specifies the size of the display area 211 of the display device 21 (step S51 in FIG. 11). And the maximum scale which can display an action area on the display area 211 is calculated (step S52 of FIG. 11). Then, the display of the display device 21 in the action area is controlled so that the display scale becomes the calculated maximum scale. Then, the display scale calculation process ends. As the maximum scale, a scale that can display the entire area of the action area and that can display the entire area of the action area is selected from scales that are predetermined for displaying the map information. For example, as a scale, values such as 1 / 100,000, 1 / 5,000, 1 / 10,000, 1 / 10,000, 1 / 5,000, etc. are predetermined and selected from these scales. The map is controlled based on the obtained value. It is not easy to display a map at a scale other than a predetermined scale. Also, if the scale is reduced, the range on the map that can be displayed is widened, but detailed information cannot be displayed on the map. Conversely, if the scale is increased, detailed information can be displayed on the map. On the other hand, the range on the map that can be displayed is narrowed. Therefore, by selecting an appropriate maximum scale, the action area can be displayed on the map, and more additional information can be displayed together with the display.

  That is, as shown in FIG. 12, the driving assistance ECU 11 specifies the size of the display area 211 of the display device 21, and narrows the first section area that constitutes the action area in the area of the display area 211. The display scale is controlled so that the narrow area C31 and the narrow area C34 obtained by narrowing the second divided area are displayed as large as possible.

  Next, the display range setting process (step S60 in FIG. 2) will be described with reference to FIGS. The display range setting process is performed on the entire action area.

  When the display range setting process is started, the driving support ECU 11 determines whether or not there is a movable range in the display area 211 of the display device 21 (step S61 in FIG. 13). The fact that there is a movable range is that the map range can be moved while displaying all the divided areas as a predetermined number or more included in the action area on the display device 21 under the calculated display scale. To be judged. When it is determined that there is no movable range in the display area 211 of the display device 21 (NO in step S61 in FIG. 13), the display range setting process is terminated because it is not necessary to change the display range setting. On the other hand, when it is determined that there is a movable range in the display area 211 of the display device 21 (YES in step S61 in FIG. 13), the driving support ECU 11 moves the display of the action area within the movable range to the display area 211. It is determined whether or not other divided areas can be included (step S63 in FIG. 13). At this time, the other divided areas are divided areas that are not included in the action area. If it is determined that the display area 211 can include other divided areas (YES in step S63 in FIG. 13), the driving support ECU 11 includes an area including the destination points of the other divided areas included in the display area 211. The display range of the action area in the display area 211 is calculated so that the probability PR becomes the maximum value (step S64 in FIG. 13).

  On the other hand, when it is determined that other display areas cannot be included in the display area (NO in step S63), or when the display range of the action area in the display area 211 is calculated (step S64), the driving assistance ECU 11 Then, it is determined whether or not the display area 211 of the display device 21 has a movable range (step S65 in FIG. 13). When it is possible to move the map range while displaying the action area on the display device 21 or displaying the action area display range calculated in step S64 under the calculated display scale, It is determined that there is a movable range. When it is determined that there is a movable range (YES in step S65 in FIG. 13), the driving assistance ECU 11 causes the display area 211 to maximize the number of other past destination points Tg that can be included in the display area 211. The display range of the action area is calculated (step S66 in FIG. 13). When it is determined that there is no movable range (NO in step S65 in FIG. 13), or when the display range of the action area in the display area 211 is calculated (step S66 in FIG. 13), the driving assistance ECU 11 displays The display range in the display area 211 of the device 21 is set (step S67 in FIG. 13). Thereby, the display range in the display device 21 is set and controlled in the action area. Then, the display range setting process ends.

  That is, as shown in FIG. 14, the driving assistance ECU 11 determines whether or not there is a movable range when the two narrow areas C31 and C34 are displayed as action areas on the display area 211 of the display device 21. First, in the drawing, the range of the action area is the right end ER at the right end position, the left end EL at the left end position, the upper end EU at the upper end position, and the lower end ED at the lower end position. At this time, the display area 211 displays the action area and can move in the figure as the distance that can be moved in the right direction SR1, the left direction SL1, the upper direction SU1, the upper direction SU1, the lower direction SD1. Respectively. That is, the maximum movable range 212 is a range in which the display range can be set. In the present embodiment, for convenience of explanation, it is assumed that no section area or past destination point Tg is set other than the display area 211.

  First, in the display area 211, the second, third and third divided areas other than the first and fourth divided areas C21 and C24 which are the basis of the two narrow areas C31 and C34 included in the action area. The five divided areas C22, C23, and C25 are defined as other areas. Then, a position at which the other area can be displayed in the display area 211 so as to have the highest probability PR of becoming an area including a new destination point is calculated. For example, it is first determined that the display area 211 can include the fifth divided area C25 having the highest probability PR of including the new destination point in the other divided areas, A display range in which five divided areas C25 are displayed is calculated. Subsequently, it is determined that the second PR area C22 having the next highest probability PR of the display area 211 including the new destination point can be included in the other segment areas. And the display range which can display 2nd division area C25, C22 is calculated. Subsequently, it is determined that the display area 211 can include the third divided area C23 having the next highest probability PR of including the new destination point in the other divided areas. 5. A display range in which the second and third divided areas C25, C22, and C23 can be displayed is calculated.

  Subsequently, in this way, it is determined whether or not there is a movable range with respect to the calculated display range. At this time, the display area 211 displays the action area and can move in the drawing as a distance that can be moved in the right direction SR2, the left distance SL2, the upper distance SU2, the upper distance SU2, the lower distance SD2 in the lower direction. Respectively. Therefore, the position where the past destination point Tg that is not included in the action area and the second, third, and fifth divided areas C22, C23, and C25 displayed in the display area 211 is the largest is calculated. In the present embodiment, since the past destination point Tg is not included in the movable direction, the display range is not changed.

  In addition, when displaying the action area and the second, third, and fifth division areas C22, C23, and C25, adjustments are made so that the margins provided on the top, bottom, left, and right are substantially equal to the area that must be displayed To do. Accordingly, the action area and the second, third, and fifth divided areas C22, C23, and C25 are displayed in an easy-to-understand manner in the display area 211 of the display device 21.

  As described above, according to the action area prediction apparatus according to the present embodiment, the effects listed below can be obtained.

  (1) A plurality of divisions selected so that the probability PR that the area size of the action area is an area including a new destination point in each of the first to fifth division areas C21 to C25 is equal to or greater than a probability threshold. Identified based on area. Thereby, it becomes possible to predict the action area not including the past destination point Tg but including the area including the new destination point. Therefore, it becomes possible to predict the action area more appropriately while using the past action history (the destination point) and suppressing the restrictions on the past action history (the destination point). For example, when going to a meal, even if the destination point is not specified, an action area whose size is specified so as to include an area including a point where it is customary to go to a meal will be predicted. . That is, even when the destination point is not clearly defined, it is possible to specify an action area including an area having characteristics expected by the driver.

  (2) Since the probability that the past destination point Tg becomes a new destination point is calculated using the Bayesian network, the probability PR that becomes the new destination point is set to “time”, “day of the week” and “departure point”. It can be suitably calculated based on various events. In addition, since the degree of freedom for an event that can be used as a cause increases, it is expected that the probability of becoming a new destination point can be appropriately calculated.

  (3) It is expected that the calculation accuracy of the probability of becoming a new destination point can be improved by using parameters highly relevant to the movement of “time”, “day of the week”, and “departure point”.

  (4) The predicted (identified) action area can be narrowed down to an area including a partitioned area with a high probability PR to be an area including a new destination point. Thereby, the probability PR of becoming an area including a new destination point in the action area can be increased, and the section area selected as the action area can be narrowed down.

  (5) Since more past destination points Tg can be included in the segmented area, the possibility that the segmented area can be partitioned as an area having characteristics expected by the driver is increased.

  (6) The divided areas included in the action area are reset so that the area becomes narrower in accordance with the number of times of visits to the destination point Tg in the past. That is, the divided area is narrowed, and the narrowed divided area is set as the narrow area. For example, if the divided area is circular, the radius r1 is reset based on the radii rx and ry, which are the distances from the center of the divided area to the destination point where the number of stops reaches the number-of-times threshold. It is reset so that the area becomes smaller.

  In addition, since the action area can be narrowed (the area of the action area can be reduced), information related to driving support can be provided more appropriately. For example, if the area of the action area is small, the scale can be increased in the map display, so that more information can be provided.

  (7) Since a predetermined number or more, for example, all of the predicted action areas are displayed in the display area 211, information related to driving assistance related to the predicted action area can be suitably provided to the driver. become. In addition, when the action area is displayed on the display device 21 in this way, it is possible to provide information on a partitioned area with a high probability PR that is an area including a new destination point, that is, information on the action area at an earlier stage.

  (8) Since it is possible to overlook an area with a high probability of being a route and it is possible to provide information about such an area, the driver is expected to reduce the scale change work. The convenience for the user is improved.

  (9) In the display range (display area 211), by displaying together with the predicted action area, an area that has a high probability of becoming a new destination point out of areas not included in the action area, The probability of including a new destination point in the display range is increased.

  (10) The probability that the display range (display area 211) includes a new destination point in the display range is increased by including more past destination points Tg together with the predicted action area and the like. .

(Other embodiments)
In addition, the said embodiment can also be implemented with the following aspects.

  In the above embodiment, the case where the information center 30 is provided has been illustrated. However, the present invention is not limited to this, and an information center may not be provided. Further, when the information center is not used, the driving assistance ECU may not be provided with a communication device. Thereby, the improvement of the design freedom degree of an action area prediction apparatus comes to be aimed at.

  In the above embodiment, the driving support ECU 11 is illustrated as a case where the navigation device 20, the in-vehicle communication device 27, the display device 21, the speaker 22, the GPS device 28, and the vehicle speed sensor 29 are connected by an in-vehicle network such as CAN. However, the present invention is not limited to this, and if necessary information communication is possible, the driving support ECU may be composed of different types and different lines of communication methods and communication lines between the devices.

  In the above embodiment, the case where the connection between the driving support ECU 11 and the navigation device 20, the in-vehicle communication device 27, the display device 21, the speaker 22, the GPS device 28, and the vehicle speed sensor 29 is CAN is illustrated. However, the present invention is not limited to this, and if necessary information can be exchanged, part or all of the connection between the driving assistance ECU and the navigation device, the in-vehicle communication device, the display screen, the speaker, the GPS device, and the vehicle speed sensor is CAN. You may be connected by other connection methods other than. Other connection methods include in-vehicle networks such as LIN, FlexRay (registered trademark), and Ethernet (registered trademark), and connection methods based on protocols thereof. Thereby, the improvement of the design freedom of an action area prediction apparatus and the expansion of an application range come to be aimed at.

  -In above-mentioned embodiment, the case where the navigation apparatus 20 and the display apparatus 21 were provided separately was illustrated. However, the present invention is not limited thereto, and a display screen that can display driving support information may be provided in the navigation device. That is, it is only necessary that information from the driving support ECU can be displayed on the display screen. Thereby, the improvement of the design freedom degree of an action area prediction apparatus comes to be aimed at.

  In the above-described embodiment, the driving assistance ECU 11 includes the divided area placement unit 111, the probability calculation unit 112, the prediction unit 113, the narrowing processing unit 114, the display scale calculation unit 115, and the display range setting unit 116. It illustrated about the case provided with. However, the present invention is not limited to this, and the divided area placement unit, the probability calculation unit, the prediction unit, the narrowing processing unit, the display scale calculation unit, and the display range setting unit can transmit and receive necessary information to a plurality of ECUs. It may be provided in a distributed manner.

  -Not limited to this, all or a part of the divided area arrangement unit, the probability calculation unit, the prediction unit, the narrowing processing unit, the display scale calculation unit, and the display range setting unit is a portable information processing device other than a vehicle. It may be provided.

  Thereby, the improvement of the freedom degree of a structure as an action area prediction apparatus is achieved.

  In the above embodiment, the driving assistance ECU 11 places the processes in the segment area placement unit 111, the probability calculation unit 112, the prediction unit 113, the narrowing processing unit 114, the display scale calculation unit 115, and the display range setting unit 116. The case of making it illustrated was illustrated. However, the present invention is not limited to this, and if each required process is executed, each unit that performs these processes includes a divided area placement unit, a probability calculation unit, a prediction unit, a narrowing processing unit, a display scale calculation unit, It does not have to be divided by division like the display range setting section. For example, a part or all of the probability calculation unit, the prediction unit, the narrowing processing unit, the display scale calculation unit, and the display range setting unit may be included in the divided area arrangement unit. Further, for example, the narrowed area processing unit and its processing may be included in the divided area arrangement unit. Thereby, the freedom degree of design as an action area prediction apparatus comes to be raised.

  In the above-described embodiment, the case where the display range is calculated so that the past destination points Tg are included the most after the other divided areas are included in the display region 211 has been illustrated. However, the present invention is not limited to this, and as long as the action area is displayed in at least the display area, only the processing for calculating the display range including the other divided areas in the display area may be performed, and the past destination points are the largest. Only the process of calculating the display range may be performed. Thereby, the improvement of the design freedom as an action area prediction apparatus comes to be aimed at.

  In the above embodiment, the case where the display range is calculated so that the number of other past destination points Tg included in the display area 211 is maximized has been illustrated. However, the present invention is not limited to this, and the display range may be calculated in a manner in which the cumulative value of the probability of becoming a new destination point is increased for past destination points included in the display area. In addition, for past destination points included in the display area, the display range may be calculated so that the cumulative value of the number of visits to the destination is increased. Thereby, the improvement of the design freedom as an action area prediction apparatus comes to be aimed at.

  In the above-described embodiment, the display range setting process includes a process for calculating a display range in which other segment areas are included in the display area 211, and a process for calculating a display range in which past destination points Tg are the largest. An example of the case is shown. However, the present invention is not limited to this, and at least if the action area is displayed in the display area, the process of calculating the display range so that the current point PC or the action area is in the center of the display area 211, or the current point PC and the action area A process for calculating the display range so that “” is in the center of the display area 211 may be included. Thereby, the improvement of the design freedom as an action area prediction apparatus comes to be aimed at.

  In the above embodiment, the case where the current location PC is included in the display area 211 is illustrated. However, the present point is not limited thereto, and the current location may not be included in the display area, for example, priority is given to a map and information display regarding the action area. Thereby, the convenience as a behavior area prediction apparatus is improved.

  In the above-described embodiment, the case where the display range setting process is performed is illustrated. However, the present invention is not limited thereto, and the display range setting process may not be performed as long as the action area is displayed at least in the display area. Thereby, the improvement of the design freedom as an action area prediction apparatus comes to be aimed at.

  -In above-mentioned embodiment, the case where the narrowing of a division area is performed after an action area prediction process was illustrated. However, the present invention is not limited to this, and the narrowing of the divided area may be performed prior to the action area prediction process. That is, the size of the action area is changed according to the change of the size of the divided area. Thereby, the improvement of the design freedom as an action area prediction process comes to be aimed at.

  In the above embodiment, the case where the display range is set for the action area that has been narrowed is illustrated. However, the present invention is not limited to this, and a display range may be set for an action area that is not narrowed. That is, even if the narrowing process is not performed on the action area, the display range can be set for such an action area. Thereby, the expansion of the application range as an action area prediction apparatus comes to be aimed at.

  In the above embodiment, the case where the display scale is calculated for the action area that has been narrowed is illustrated. However, the present invention is not limited to this, and the display scale may be calculated according to an action area that is not narrowed. That is, even if the narrowing process is not performed on the action area, the display scale can be calculated for the action area. Thereby, the expansion of the application range as an action area prediction apparatus comes to be aimed at.

  -In above-mentioned embodiment, it illustrated about the case where the narrowing-down process of an action area is performed. However, the present invention is not limited to this, and the action area narrowing process may not be performed. Even if the narrowing process is not performed, the action area is appropriately predicted because the action area is composed of segmented areas with a high probability of becoming a new destination point. Thereby, the freedom degree of design as an action area prediction apparatus comes to be raised.

  In the above embodiment, the case where the divided area to be narrowed is circular is illustrated. However, the present invention is not limited to this, and the divided area to be narrowed may have any shape such as an ellipse, a rectangle, a polygon such as a hexagon.

  For example, if the segmented area is a rectangle (rectangle), the length of each side is the distance from the center of the segmented area to the destination point where the number of stops reaches the number of times threshold, and the distance to the side on the extension. By resetting based on the ratio, the resetting is performed so that the divided area becomes narrower. That is, the area of the area can be narrowed down, that is, the size of the area can be reduced while maintaining a high probability that the action area is an area including a new destination point.

  In the above embodiment, the case where the number-of-times threshold is compared with the “cumulative number of visits / total number of visits” is illustrated. However, the present invention is not limited to this, and if the probability of becoming an area including a new destination point in the divided area can be maintained high, for example, the number threshold may be set by the number of times and compared with the cumulative number of stops. Thereby, the improvement of the design freedom degree of an action area prediction apparatus comes to be aimed at.

  In the above embodiment, the case where the divided area is narrowed (smaller) is illustrated. However, the present invention is not limited to this, and the size of the area of the divided area may be increased to increase the probability of becoming an area including a new destination point. Thereby, the improvement of the design freedom degree of an action area prediction apparatus comes to be aimed at.

  -In the said embodiment, the 1st-5th division area C21-C25 is the past objective which includes the 1st-5th division area C11-C15 temporarily arrange | positioned around the vehicle 10 at equal intervals there The case where the arrangement position is determined based on the process of moving so as to increase the number of points, that is, the so-called cluster process is illustrated. However, the present invention is not limited to this, and as long as the arrangement position of the divided area can be suitably determined, the arrangement position of the divided area may be determined by other cluster processing. For example, cluster processing may be performed so as to collect places where the past destination points are densely packed so as to obtain a predetermined area. Thereby, the freedom degree of design as an action area prediction apparatus comes to be raised.

  -In the above-mentioned embodiment, it illustrated about the case where the action area was predicted based on the 1st-5th division area C21-C22 which rearranged the 1st-5th division area C11-C12 temporarily arranged. . However, the present invention is not limited to this, and the action area may be predicted based on the temporarily arranged first to fifth divided areas. Thereby, simplification of the structure of an action area prediction apparatus comes to be achieved.

  In the above embodiment, the case where the input values of the Bayesian network are “time”, “day of the week”, “departure point”, and “destination point” is exemplified. However, the present invention is not limited to this, and if the probability that a new destination point can be calculated for at least a past destination point can be calculated, the input value is “time”, “day of the week” except for “destination point”. And at least one of “departure points” may be included, and other elements other than these may be included. Alternatively, it may be other elements other than “time”, “day of the week”, and “departure point”. For example, other factors include weather, brightness, driver, passengers, number of passengers, loading weight, and the like. Thereby, the freedom degree of design and applicability as an action area prediction apparatus come to be improved.

  In the above embodiment, the case where the probability estimation process is performed using a Bayesian network is illustrated. However, the present invention is not limited to this, and the probability estimation process may be performed by other algorithms as long as the probability that a past destination point becomes a new destination point can be calculated based on the past action history. Good. For example, other algorithms include those using a neural network, those using a tree diagram, and those using a route. Information necessary for other algorithms may be acquired from history information or the like. Thereby, the freedom degree of design as an action area prediction apparatus comes to be raised.

  In the above embodiment, the probability PR of becoming an area including a new destination point corresponding to one division area is added to the probability of being a new destination point of all past destination points included in the division area. It illustrated about the case where it calculates by. However, the present invention is not limited to this, and if it is possible to appropriately compare the probability of becoming an area including a new destination point, for example, the probability of becoming a new destination point is averaged or statistical processing is performed. The probability of becoming an area including a new destination point may be calculated by a calculation method other than addition, such as. Thereby, the freedom degree of design as an action area prediction apparatus comes to be raised.

  In the above embodiment, the case where the probability PR of an area including a new destination point is indicated as a percentage is illustrated. However, the present invention is not limited to this, and information such as a numerical value indicating a ratio or a numerical value indicating a magnitude relationship may be used as the probability of becoming an area including a new destination point as long as it is numerical information that can be used for specifying the ranking. Thereby, the freedom degree of design as an action area prediction apparatus comes to be raised.

  -In above-mentioned embodiment, it illustrated about the case where the surrounding area of the vehicle 10 was made into the range of radius 10km centering on present position PC. However, the size and shape of the surrounding area is not limited to this, and if it is expected that a divided area with a high probability of becoming an area including a new destination point can be obtained, a circular or elliptical shape with a radius other than 10 km is used. Any shape such as a fan shape, a rectangle, and other polygonal shapes may be used, and a large number of traveling directions of the vehicle may be included. Thereby, the freedom degree of design as an action area prediction apparatus comes to be raised.

  -In above-mentioned embodiment, the temporary center points P11-P15 illustrated about the case where it arrange | positions at intervals of 4 km in the east-west-north-north (up-down, left-right in a figure) direction, respectively. However, the present invention is not limited to this, and the interval at which the temporary center point is arranged may be other than 4 km as long as the divided area including the temporary center point can be appropriately temporarily arranged in the surrounding area. A plurality of intervals may be employed. The interval at which the temporary center points are arranged may be such that the arrangement is shifted in at least one of the vertical direction and the horizontal direction, is arranged according to other rules, or is random. Thereby, the freedom degree of design as an action area prediction apparatus comes to be raised.

  -In above-mentioned embodiment, it illustrated about the case where the 1st-5th division area C11-C15 temporarily arrange | positioned is circular with a radius of 2 km. However, the present invention is not limited to this, and the size and shape of the segmented area may be a circle with a radius other than 2 km, an ellipse, a rectangle, a hexagon, etc. Any shape such as a polygon may be used. The center point when it is not circular may be appropriately defined, and it is possible to obtain a similar shape that increases or decreases the area of the area regardless of the shape. Thereby, the freedom degree of design as an action area prediction apparatus comes to be raised.

  -In above-mentioned embodiment, the case where the action area prediction apparatus was used for the vehicle 10 was illustrated. However, the present invention is not limited to this, and the action area prediction apparatus may be used for a moving body other than a vehicle, such as a ship or a robot. Moreover, the action area prediction device may be used for a portable information processing device such as a smartphone. Furthermore, the action area prediction apparatus may be provided in an information center or the like. In this case, it is only necessary that information such as an action area predicted by an information center can be transmitted to an in-vehicle device or a portable information processing device in which the information area is used. Thereby, the application range of the action area prediction apparatus can be expanded.

  DESCRIPTION OF SYMBOLS 10 ... Vehicle, 11 ... Driving assistance ECU, 20 ... Navigation device, 21 ... Display device, 22 ... Speaker, 27 ... In-vehicle communication device, 28 ... GPS device, 29 ... Vehicle speed sensor, 30 ... Information center, 31 ... Center communication device , 32 ... Information providing unit, 33 ... Traffic information DB, 34 ... Weather information DB, 35 ... Map information DB, 111 ... Segmental area arranging unit, 112 ... Probability calculating unit, 113 ... Predicting unit, 114 ... Narrowing processing unit , 115 ... display scale calculation unit, 116 ... display range setting unit, 117 ... movement history management unit, 211 ... display area, 121 ... action history DB, 122 ... traffic history DB, 123 ... map information DB, PC ... current location, PR: Probability of becoming an area including a new destination point, Tg: Past destination point, C11 to C12 ... Temporarily arranged first to fifth divided areas, C21 to C25 ... First to fifth Segment area, C31 ... narrow area (first section area), C34 ... narrow area (fourth division area), P11~P15 ... temporary center point, P21~P25 ... center point.

Claims (10)

A behavior area prediction device that predicts a behavior area including an area that is a destination based on past behavior history information associated with map information,
A divided area arrangement unit for arranging a divided area, which is an area having a predetermined size, on a map of map information corresponding to the past action history information;
For each segment area, the probability that a past destination point included in the segment area becomes a new destination point is calculated, and the probability that the segment area becomes a new destination point based on the calculated probability A probability calculation unit for calculating
The size of the area of the action area is specified in a size including one or a plurality of divided areas selected so that the probability of becoming an area including a new destination point is not less than a predetermined value from the plurality of divided areas Specific part to do,
A behavior area prediction apparatus comprising: The action area prediction device according to claim 1, wherein the specifying unit sequentially selects the action area from a segment area having a high probability of becoming an area including a new destination point. The divided area arrangement unit sets the divided area as a circular area, and when the divided area is arranged, the past is the largest in the range including the center points of the divided areas arranged at equal intervals on the map. The action area prediction device according to claim 1 or 2, wherein the divided areas are rearranged at positions including the destination point. Each of the behavior history information is associated with information indicating the number of times the user visited the destination point in the past,
The identification unit accumulates the number of stoppages of the destination points included in the divided area included in the action area in the order from the closest to the center of the divided area, and the accumulated number of stoppages reaches the stop number threshold. The action area prediction apparatus according to any one of claims 1 to 3, wherein the size of the divided area is reset based on a distance from the destination to the destination point. In the action area prediction device according to any one of claims 1 to 4,
A display device capable of displaying an image of the map information and the action area together;
A behavior area prediction device, comprising: a display control unit that controls a display scale of the map information so that a predetermined number or more of divided areas included in the behavior area are displayed on the display device. The action area prediction apparatus according to claim 5, wherein the display control unit further controls the display scale so that a current location is also displayed. The display control unit is included in the action area when moving the map range while displaying a predetermined number or more of the divided areas included in the action area on the display device under the controlled display scale. The action area prediction apparatus according to claim 5 or 6, wherein the display range is set and controlled in such a manner that an area with a high probability of including a new destination point is displayed together even if the area is not a divided area. The behavior area prediction apparatus according to claim 7, wherein the display control unit further controls the display range in a manner in which a large number of past destination points not included in the behavior area are included. The behavior area prediction apparatus according to any one of claims 1 to 8, wherein the probability calculation unit calculates a probability that a past destination point becomes a new destination point using a Bayesian network. The behavior area prediction apparatus according to claim 9, wherein the probability calculation unit uses the Bayesian network so as to include at least one of a time, a day of the week, and a departure point as an event that becomes a “cause” of the Bayesian network.

Images