JP5169556B2 - Destination prediction apparatus and destination prediction method - Google Patents

Description

  The present invention relates to a destination prediction apparatus and a destination prediction method for predicting a destination of a vehicle.

Conventionally, as a technique for predicting the destination of a vehicle, information such as the current vehicle position and day of the week is detected and stored in a predetermined travel history storage means as the engine ignition switch is turned on. Based on the history of past departure place, arrival place, own vehicle position, day of the week, etc., the arrival place when the current own vehicle position and day of the week are the same is extracted, and the number of arrivals is the most among the extracted arrival places A navigation device that predicts many items as destinations is known (see Patent Document 1).
JP 2005-292029 A

  In the conventional technology, since the destination with the highest number of arrivals is predicted as the destination in the driving history, it depends on the day of the week or the time zone, such as going to a beauty salon once a month. However, it is difficult to predict destinations that are likely to occur regularly even if they are infrequent, and the prediction accuracy of destinations that can be provided to users may be insufficient. It was happening.

  The present invention has been proposed in view of the above circumstances, and provides a destination prediction apparatus and a destination prediction method capable of accurately predicting a destination having a high probability of periodically occurring. There is.

  The present invention stores the departure / arrival history from the departure point to the arrival point in the past traveling of the host vehicle in a predetermined traveling history storage means, and calculates the visit interval for each arrival point included in the stored departure / arrival history. Then, based on the calculated visit interval for each arrival place, the presence or absence of regularity is determined for each arrival place. Of the arrival places determined to have regularity, the elapsed time from the last visit to the present If there is an arrival place that matches the calculated visit interval, the arrival place is predicted as a destination and presented to the user.

  According to the present invention, even if the frequency is low, it is possible to detect an action that has a high probability of occurring regularly, and therefore the destination of the action can be accurately predicted and presented to the user. Can do.

  Hereinafter, a destination prediction apparatus as a preferred embodiment of the present invention will be specifically described.

[Configuration of destination prediction device]
As shown in FIG. 1, the destination prediction apparatus shown as an embodiment of the present invention includes a GPS signal receiving unit 1 that receives a GPS (Global Positioning System) signal, and a driving start that detects the start and end of driving of the host vehicle. End detection unit 2, travel history storage unit 3 that stores past travel history of the host vehicle, map database 4 that stores map data, facility information database 5 that stores facility information, and the destination prediction device A control unit 6 that performs overall control, an information display unit 7 that displays various types of information including map data and facility information, and an information input unit 8 that inputs various types of information.

  The GPS signal receiving unit 1 receives a GPS signal from a GPS satellite in order to specify the current date and time, the date, and the day of the week in addition to the current position of the host vehicle. The GPS signal receiving unit 1 supplies the control unit 6 with information such as the current position, date / time, month / day, day of the week, and the like based on the received GPS signal. The driving start / end detection unit 2 detects the start and end of driving of the host vehicle by detecting the state of the ignition key. When the operation start / end detection unit 2 detects the start and end of the operation, the operation start / end detection unit 2 supplies detection information indicating the fact to the control unit 6.

  The travel history storage unit 3 is configured using, for example, a RAM (Random Access Memory) or the like, and as shown in FIG. As shown in FIG. 3, the names of departure and arrival locations and latitude and longitude are stored as a registration point database together with the departure and arrival history in traveling. Information stored in the travel history storage unit 3 is read by the control unit 6.

  Each of the map database 4 and the facility information database 5 is configured by using a general storage medium such as a cache memory, a main memory, a hard disk, a CD, an MD, a DVD, an optical disk, or an FDD, for example. Stores map data including road information (link information) and facility information (node information) used when providing guidance. The map data and facility information stored in the map database 4 and the facility information database 5 are read out by the control unit 6, respectively.

  The control unit 6 is, for example, a general processing unit such as a CPU (Central Processing Unit), an MPU (Micro Processing Unit), a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field Programmable Gate Array). Based on information read from the driving history storage unit 3, the map database 4, and the facility information database 5, which is configured by combining operation circuits and supplied from the GPS signal receiving unit 1 and the operation start / end detection unit 2, Centrally control the destination prediction apparatus. The configuration of the control unit 6 will be described later.

  The information display unit 7 is configured using, for example, a CRT (Cathode Ray Tube), a liquid crystal display, or the like. When performing route guidance for the host vehicle, the information display unit 7 is controlled by the control unit 6 and the map database 4 or the facility information database. 5 displays various information such as map data and facility information stored therein, and presents the route to the destination to the driver.

  The information input unit 8 is configured using an input device such as a touch panel, a remote controller, a mechanical switch, or a voice input device, for example. Under the control of the control unit 6, an output message or a presented list It is provided for the user to input a reply to. Information input via the information input unit 8 is supplied to the control unit 6.

[Configuration of control unit]
The control unit 6 stores the visit interval calculation unit 61 that calculates the visit interval for each arrival place, the regularity presence / absence determination unit 62 that determines whether or not each arrival place has regularity, and the travel history storage unit 3. A departure / arrival history search unit 63 that searches for a departure / arrival history, a destination prediction unit 64 that predicts a destination from arrival points included in the arrival / departure history searched by the departure / arrival history search unit 63, and an operation start / end detection unit 2 includes a point name search unit 65 that searches for the names of points when the start and end of driving are detected, and a route calculation unit 66 that calculates a route to the destination.

  The visit interval calculation unit 61 calculates a visit interval for each arrival place based on the departure / arrival history stored in the travel history storage unit 3. The visit interval calculation unit 61 supplies information indicating the calculated visit interval to the periodicity presence / absence determination unit 62 and the destination prediction unit 64. The periodicity presence / absence determining unit 62 determines the presence / absence of regularity at each arrival location based on the visit interval for each arrival location calculated by the visit interval calculating unit 61. The periodicity presence / absence determining unit 62 supplies the destination predicting unit 64 with information indicating the arrival place determined to have periodicity.

  The departure / arrival history search unit 63 includes information on the current position, date / time, month / day, day of the week, etc. supplied from the GPS signal receiving unit 1 and the departure when the driving start / end detecting unit 2 detects the start of driving. Using the position as a search condition, a travel history that matches the search condition is searched from the travel history storage unit 3. The arrival / departure history search unit 63 supplies information indicating the search result to the destination prediction unit 64.

  The destination predicting unit 64 determines, from the time of the last visit, the arrival locations determined by the periodicity determination unit 62 among the arrival locations included in the arrival / departure history searched by the arrival / departure history search unit 63. The elapsed time up to the present time is compared with the visit interval calculated by the visit interval calculation unit 61, and when there is an arrival place where the elapsed time and the visit interval substantially coincide with each other, Predict as the ground. The destination prediction unit 64 supplies information indicating the predicted destination to the route calculation unit 66.

  The point name search unit 65 is the name of a point when the point at which the start and end of driving is detected by the driving start / end detection unit 2 is not stored in the registered point database in the travel history storage unit 3. Is searched from the point where the start and end of driving are detected. The point name search part 65 registers the name selected via the information input part 8 among the names of the searched points in the registration point database as the name of the point.

  When the destination is predicted by the destination prediction unit 64, the route calculation unit 66 calculates a route to the destination. The route calculation unit 66 presents information indicating the calculated route to the driver via the information display unit 7.

[Operation of destination prediction device]
Such a destination prediction apparatus predicts a destination in consideration of not only the day of the week and the time zone, but also the regularity of the destination.

  For example, based on the arrival / departure history as shown in FIG. 4, consider a case of predicting the destination when “departing home on Sunday, December 22”. In the conventional destination prediction apparatus, when the destination is predicted based on the arrival / departure history, the supermarket will search 10 times when searching for the arrival place where “Departure = Home” and “Day of the Week = Sunday”. Because the pet shop will be 5 times, the company will be 5 times, and the hair salon will be 4 times, it is predicted that this is the destination with the highest probability of visiting the supermarket with the highest number of visits. However, when calculating the interval between visits to the beauty salon, it is all 6 weeks, and considering that the elapsed time from the last visit to the beauty salon to December 22 is exactly 6 weeks, It can be seen that the person visits the beauty salon regularly, and now that it has been 6 weeks since the last visit, the possibility of going to the beauty salon is high.

  In the destination prediction apparatus shown as an embodiment of the present invention, a purpose that considers the regularity inherent in each destination different from the day of the week and time zone, which could not be handled by the conventional destination prediction apparatus Perform land prediction.

[Departure and Arrival History Processing]
First, the departure / arrival history storage process for the travel history storage unit 3 will be described.

  In the destination prediction apparatus, as shown in FIG. 2, the driving history storage unit 3 uses the driving start date / time, day of the week, driving start point, and driving end point as arrival / departure history whenever the host vehicle is driven. 3, and as shown in FIG. 3, the latitude and longitude of the points where the driving has started and ended in the past and the point names assigned thereto are accumulated and stored in the driving history storage unit 3 as a registered point database.

  In addition, in the destination prediction apparatus, when departure and arrival are made at a point not stored in the registered point database, the point name search unit 65 determines a point name close to the latitude and longitude of the point in the facility information database. 5 and is presented as a candidate name to the user via the information display unit 7. In response to this, the user selects the name of a desired spot from the candidate names presented in the list via the information input unit 8. Thereby, in the destination prediction apparatus, the name selected via the information input part 8 can be registered into the registration spot database by the spot name search part 65 as a name of a new spot.

  The destination prediction apparatus can perform destination prediction by storing the departure / arrival history and the registration point database in the travel history storage unit 3 in this manner. Specifically, in the destination prediction apparatus, the following periodicity determination process is performed to determine whether each arrival point included in the arrival / departure history has periodicity, and based on the result, the destination Perform prediction processing.

[Regularity presence / absence discrimination processing]
Specifically, the destination prediction apparatus determines the presence or absence of regularity at each destination included in the departure / arrival history stored in the travel history storage unit 3 according to a series of procedures as shown in FIG.

  First, in the destination prediction device, as shown in FIG. 5, the visit interval calculation unit 61 determines whether or not a point is registered in the registration point database in step S1. Here, when there is no registered point, the destination predicting apparatus ends the series of periodicity determination processing as it is. On the other hand, when there is a registered point, the destination prediction apparatus determines whether or not there is still periodicity among the points registered in the registered point database by the visit interval calculation unit 61 in step S2. Select one point that has not gone.

  Subsequently, in step S3, the visit interval calculation unit 61 refers to the departure / arrival history stored in the travel history storage unit 3, and determines whether or not the site selected in step S2 has been visited three times or more. Determine. Note that the number of visits is three or more. If the number of visits is two or less, only one visit interval is calculated, and two or more visit intervals cannot be compared. This is because it cannot be determined. Therefore, the visit interval calculation unit 61 sets a lower limit of 3 times as a threshold value for the number of visits. However, when determining periodicity with higher accuracy, the number of visits may be four or more, instead of three. If the destination prediction apparatus has not visited three or more times, the process proceeds to step S8. If the destination prediction apparatus has visited three or more times, the process proceeds to step S4.

  When the destination prediction apparatus shifts the process to step S4, the visit interval calculation unit 61 calculates all the visit intervals of the points selected in step S2. At this time, the visit interval calculation unit 61 may calculate in units of days, or may calculate in units of hours, minutes, or seconds.

  Subsequently, in step S5, the destination prediction apparatus calculates the average value and standard deviation of the visit intervals using the visit interval calculated in step S4 by the visit interval calculation unit 61. The destination prediction apparatus determines the presence or absence of regularity based on the standard deviation calculated here, that is, the magnitude of the variation in visit interval.

  That is, in step S6, the destination prediction apparatus determines whether or not the standard deviation calculated in step S5 is 2 days or less by the periodicity presence / absence determination unit 62, and is larger than 2 days. In this case, the process proceeds to step S8. On the other hand, if it is two days or less, it is determined in step S7 that there is periodicity. Here, the upper limit of the standard deviation is set to 2 days, but the destination prediction apparatus determines that there is periodicity only when there is no variation and all the visit intervals are the same. Alternatively, the upper limit of the standard deviation may be changed according to the average value of the visit intervals. For example, in the destination prediction apparatus, when the average value of visit intervals is less than 7 days, the standard deviation for determining that there is regularity is 0 days, and the average value of visit intervals is 8 days or more If it is less than 14 days, the standard deviation for determining that there is regularity shall be 1 day, and if the average value of visit intervals is 14 days or more and less than 21 days, it shall be regular The standard deviation is 2 days. If the average value of visit intervals is 21 days or more, the standard deviation is 7 days. May be determined to have regularity when is sufficiently smaller than the average value.

  As described above, the destination prediction apparatus can determine whether or not there is regularity at the point selected in step S2. In step S8, the destination prediction apparatus determines whether or not a point that has not yet been determined whether or not there is regularity is registered in the registered point database. The destination predicting apparatus repeats the processing from step S2 when there is a point where the presence / absence of regularity is not determined, and ends the series of periodicity determination processing when there is no such point. Then, according to a series of procedures as shown in FIG. 6, the destination prediction process using regularity is performed.

[Destination prediction process]
First, as shown in FIG. 6, the destination prediction apparatus determines whether or not the operation has started by determining whether or not the ignition key has been inserted by the operation start / end detection unit 2 in step S <b> 11. The destination prediction apparatus repeats the process until it detects that the ignition key has been inserted, and when detecting that the ignition key has been inserted, in step S12, when the ignition key is inserted by the GPS signal receiving unit 1. That is, information such as the current position (departure place) date and time, the date, and the day of the vehicle at the time of starting driving is acquired.

  Subsequently, in step S13, the destination predicting apparatus searches the registration point name close to the departure point acquired in step S12 by using the arrival / departure history search unit 63 from the registration point database in the travel history storage unit 3, The search history is stored in the travel history storage unit 3 using “is a registered spot name” as a search condition.

  Then, in step S14, the destination prediction apparatus includes the periodicity described with reference to FIG. 5 in each destination included in the arrival / departure history searched in step S13 by the destination prediction unit 64. It is determined whether or not there is a point determined to have periodicity in the determination process.

  Here, if there is no point determined to have regularity, the destination prediction apparatus shifts the processing to step S17, and the destination prediction unit 64 sets the destination with the highest number of visits. It is determined that it is the ground, and the process proceeds to step S18.

On the other hand, when there are points determined to have regularity, the destination predicting apparatus determines each of the points having regularity determined in step S14 by the destination predicting unit 64 in step S15. The periodicity evaluation value P is calculated as an index representing the probability of a visit to each destination. Specifically, the destination prediction unit 64 calculates the periodicity evaluation value P as follows. FIG. 7 shows a graph in which the horizontal axis represents the time (number of days) with the origin at the time of the last visit to a certain point, and the vertical axis represents the probability that a visit to the point occurs. Further, the function f (t) in FIG. 7 is a probability density function expressed by the following formula 2. In Expression 2, μ is an average value of visit intervals at a certain point A, and σ is a standard deviation of visit intervals at the point A. T is the elapsed time from the last visit to the point A to the present.

  The destination prediction unit 64 visits the point A based on the elapsed time t and the probability density function f (t) represented by the average value μ of the visit intervals to the point A and the standard deviation σ. The probability, that is, the periodicity evaluation value P is calculated. The destination prediction unit 64 calculates the periodicity evaluation value P of all points determined to have periodicity by such a method. Subsequently, in step S16, the destination predicting device discriminates the highest value among the periodicity evaluation values P of the arrival destinations having regularity calculated in step S15 by the destination predicting unit 64. The point corresponding to the periodicity evaluation value P is determined as the destination.

  The destination prediction unit 64 does not discriminate the arrival destination having the highest periodicity evaluation value P as the destination, but the user through the information display unit 7 as a list in descending order of the periodicity evaluation value P. The arrival place selected by the user from the list may be determined as the destination. Further, the destination prediction unit 64 may determine that a point having regularity in which the elapsed time t from the last visit to the present coincides with the average value μ of the visit intervals is the destination. . Further, the destination predicting unit 64 sets the destination as the destination when the difference between the elapsed time t from the last visit to the present and the average value μ of the visit intervals is within the standard deviation σ of the visit intervals. You may make it discriminate | determine.

  Furthermore, the destination prediction unit 64 uses the probability density function f (t) to obtain the periodicity evaluation value P. However, the destination prediction unit 64 does not target the whole, but performs so-called 3-sigma processing. Those with a low probability may be excluded in advance. That is, when the difference between the elapsed time t from the last visit to the point A to the present and the average value μ of the visit intervals exceeds three times the standard deviation σ, the destination predicting unit 64 has a very high probability. Therefore, the periodicity evaluation value P may be calculated as 0. In addition, when the same periodicity evaluation value P is calculated, the destination predicting unit 64 may determine an arrival place with a large number of visits as a destination.

  When the destination predicting apparatus predicts and sets the destination in this way, in step S18, the route calculation unit 66 calculates a route from the current position to the set destination using the map database 4, In step S19, the calculated route information is presented to the user via the information display unit 7, and the series of destination prediction processes ends.

  The destination prediction apparatus determines whether or not each arrival point is included in the arrival / departure history stored in the travel history storage unit 3 according to such a series of procedures, and uses the periodicity as an index. The destination can be predicted by increasing the priority of the destination.

  As described above in detail, in the destination prediction apparatus shown as the embodiment of the present invention, the departure / arrival history from the departure point to the arrival point in the past traveling of the host vehicle is stored in the predetermined traveling history storage unit 3. The visit interval calculation unit 61 calculates the visit interval for each arrival location included in the stored arrival / departure history, and based on the calculated visit interval for each arrival location, the presence / absence of regularity of each arrival location is determined. Of the arrival places determined by the periodicity presence / absence determining unit 62 and determined to have regularity, the elapsed time from the last visit to the present and the visit interval calculated by the visit interval calculating unit 61 are substantially equal. If there is a destination that has been made, the destination prediction unit 64 predicts the destination as the destination and presents it to the user. As a result, in this destination prediction apparatus, even if the frequency is low, it is possible to detect an action that has a high probability of occurring periodically, so that the destination of such an action can also be accurately predicted and used. Can be presented to the person. Specifically, in this destination predicting apparatus, there is periodicity when the visit intervals for any destination calculated by the visit interval calculating unit 61 are all matched by the periodicity determining unit 62. By determining that the destination is a thing, it is possible to accurately predict a destination that has a high probability of periodically occurring even if the frequency is low.

  Further, in this destination prediction apparatus, the regularity presence / absence determining unit 62 calculates the standard deviation σ of the visit interval for any destination calculated by the visit interval calculating unit 61, and the calculated standard deviation σ is Whether or not the degree of variation is sufficiently small even if there is variation in visit intervals by determining that there is regularity when it is sufficiently smaller than the average value μ of visit intervals It is possible to make a determination, and it is possible to predict a destination having a high probability of occurrence periodically with higher accuracy.

  Further, in this destination predicting apparatus, an elapsed time t from the time of the last visit to an arbitrary arrival point determined by the destination predicting unit 64 as having periodicity by the periodicity determining unit 62 to the present time, The average value μ of the visit intervals calculated by the visit interval calculation unit 61 is compared, and when the difference is within the standard deviation σ of the visit interval, the arrival place is predicted as the destination, Even if there is variation in the interval, it is possible to determine whether or not the degree of variation is sufficiently small, and it is possible to predict a destination with a high probability of regularly occurring with higher accuracy. it can.

  Furthermore, in this destination predicting apparatus, the visit interval calculated by the visit interval calculating unit 61 is determined for each arrival place determined by the destination predicting unit 64 as having regularity by the periodicity determining unit 62. The probability density function f (t) with the origin at the time of the last visit to the destination represented by the above equation (1) using the average value μ and the standard deviation σ, and the time of the last visit to the destination By calculating the periodicity evaluation value P that is the probability that a visit to the destination will occur based on the elapsed time t from the present to the present, and predicting the destination based on the calculated periodicity evaluation value P, Even if there are variations in visit intervals, it is possible to determine whether the degree of variation is sufficiently small, and even if there are multiple such arrival locations, Highly accurate destinations with a high probability of visiting Can be predicted.

  Further, in this destination prediction apparatus, the destination that has the highest periodicity evaluation value P among the destinations determined by the destination prediction unit 64 as having regularity by the periodicity determination unit 62 is used as the destination. By predicting as a place, even when there are a plurality of arrival places with varying visit intervals, a destination having the highest probability of visiting can be predicted with high accuracy. Further, in this destination prediction apparatus, the destination prediction unit 64 lists the arrival destinations that are determined to have periodicity by the periodicity presence / absence determination unit 62 in order from the arrival point having the highest periodicity evaluation value P. The destination is the destination selected by the user from the list, and even if there are multiple destinations with varying visit intervals, the intention of the user is reflected. The predicted destination can be predicted with high accuracy.

  Furthermore, in this destination prediction apparatus, the elapsed time t from the time of the last visit to the present among the arrival places determined by the destination prediction unit 64 as having periodicity by the destination prediction unit 64 is calculated. Even if there is a variation in the visit interval, it is determined whether or not the degree of the variation is sufficiently small by predicting the arrival point that matches the average value μ of the visit interval as the destination. Therefore, it is possible to predict a destination having a high probability of occurrence periodically with higher accuracy.

  The above-described embodiment is an example of the present invention. For this reason, the present invention is not limited to the above-described embodiment, and even if it is a form other than this embodiment, as long as it does not depart from the technical idea according to the present invention, the design and the like Of course, various modifications are possible.

It is a block diagram shown about the structure of the destination prediction apparatus shown as embodiment of this invention. It is a figure shown about the specific example of the arrival / departure history memorize | stored in the travel history memory | storage part in the destination prediction apparatus shown as embodiment of this invention. It is a figure shown about the specific example of the registration point database memorize | stored in the travel history memory | storage part in the destination prediction apparatus shown as embodiment of this invention. It is a figure shown about the example of the arrival / departure history memorize | stored in the travel history memory | storage part in the destination prediction apparatus shown as embodiment of this invention, and is a figure for demonstrating the difference with the conventional destination prediction apparatus. It is a flowchart shown about a series of procedures at the time of performing regularity presence determination processing in the destination prediction apparatus shown as embodiment of this invention. It is a flowchart shown about a series of procedures at the time of performing the destination prediction process in the destination prediction apparatus shown as embodiment of this invention. It is a figure for demonstrating the calculation method of a periodicity evaluation value.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 GPS signal receiving part 2 Driving start / end detection part 3 Running history storage part 4 Map database 5 Facility information database 6 Control part 7 Information display part 8 Information input part 61 Visit interval calculation part 62 Periodicity presence / absence judgment part 63 Arrival / departure history search Unit 64 Destination prediction unit 65 Point name search unit 66 Route calculation unit

Claims (9)

Traveling history storage means for storing departure / arrival history from the departure place to the arrival place in the past traveling of the host vehicle;
A visit interval calculating means for calculating a visit interval for each destination included in the departure and arrival history stored in the travel history storage means;
Based on the visit interval for each arrival place calculated by the visit interval calculation means, periodicity presence / absence determination means for determining the presence / absence of regularity at each arrival place;
Among arrival places determined to be periodic by the periodicity determination means, the arrival time from the last visit to the present and the visit interval calculated by the visit interval calculation means match. A destination prediction device, comprising: a destination prediction unit that predicts the arrival location as a destination and presents it to the user when there is a location. The destination prediction apparatus according to claim 1,
The destination predicting device, wherein the periodicity presence / absence determining means determines that there is periodicity when all the visit intervals for any destination calculated by the visit interval calculating means match. . The destination prediction apparatus according to claim 1,
The periodicity presence / absence determining means calculates the standard deviation of the visit interval for any destination calculated by the visit interval calculating means, and the calculated standard deviation is sufficiently smaller than the average value of the visit intervals A destination predicting device characterized by discriminating that the device has periodicity. In the destination prediction apparatus according to claim 3,
The destination predicting means includes an elapsed time from the time of the last visit to an arbitrary arrival place determined to be periodic by the periodicity presence / absence determining means and a visit calculated by the visit interval calculating means. A destination prediction apparatus, which compares an average value of intervals and predicts the destination as a destination when the difference is within the standard deviation of the visit interval. In the destination prediction apparatus according to claim 2 or claim 3,
The destination predicting means uses an average value μ and standard deviation σ of visit intervals calculated by the visit interval calculating means for each arrival place determined to have periodicity by the periodicity determining means. Based on the probability density function f (t) with the origin at the time of the last visit to the destination as represented by the following formula, and the elapsed time t from the time of the last visit to the destination to the present, A destination prediction apparatus that calculates a periodicity evaluation value that is a probability that a visit to the ground will occur and predicts a destination based on the calculated periodicity evaluation value.

The destination prediction apparatus according to claim 5,
The destination predicting means predicts, as a destination, an arrival place having the highest periodicity evaluation value among arrival places determined to have periodicity by the periodicity presence / absence determining means. Land prediction device. The destination prediction apparatus according to claim 5,
The destination predicting means presents to the user as a list in order from the arrival place having the highest periodicity evaluation value among the arrival places determined to have periodicity by the periodicity presence / absence determining means, and from the list A destination prediction apparatus characterized in that the destination selected by the user is a destination. The destination prediction apparatus according to claim 5,
The destination predicting means has an elapsed time t from the time of the last visit to the present among the arrival places determined by the periodicity presence / absence determining means to be equal to the average value μ of the visit intervals. A destination prediction apparatus for predicting an arrival point as a destination. A travel history storage step of storing a departure / arrival history from the departure place to the arrival place in the past travel of the host vehicle in a predetermined travel history storage means;
A visit interval calculating step for calculating a visit interval for each arrival place included in the departure and arrival history stored in the travel history storage means;
Based on the visit interval for each arrival place calculated in the visit interval calculation step, a periodicity presence / absence determination step for determining the presence / absence of each arrival place,
Of the arrival places determined to have periodicity in the periodicity presence / absence determining step, the elapsed time from the last visit to the present and the visit interval calculated in the visit interval calculating step match. And a destination prediction step of predicting the destination as a destination and presenting it to the user.

Images