Apparatus and method for predicting destination

Abstract

The invention relates to an apparatus and a method for predicting a destination. An apparatus for predicting a destination includes one or more processors and a memory configured to store one or more programs executed by the one or more processors, where the processors may classify trajectory data collected from a vehicle using a networked automotive service into a plurality of travel routes, for each of a plurality of travel routes, departure information relating to a departure point and arrival information relating to an arrival point are generated. Predicting one or more travel destinations for each piece of departure information on the basis of at least one of the correlation between departure information and arrival information generated for each of the plurality of travel routes and whether or not the same travel route among the plurality of travel routes appears regularly; and based on detecting whether the destination is input, automatically providing the predicted driving destination to control an operation of the vehicle.

Description

[0001] Cross-reference to related applications

[0002] This application claims priority and benefit to Korean Patent Application No. 10-2024-0125528, filed on September 13, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present invention relates to an apparatus and method for predicting a destination, and more particularly, to an apparatus and method for predicting a destination capable of predicting a destination for each departure rule based on the vehicle's travel trajectory. Background Technology

[0004] The destination recommendation technology provided in existing vehicles offers the destinations most frequently visited by most users, and therefore does not recommend customized destinations, but rather popular points of interest (POIs).

[0005] Therefore, in order to recommend customized destinations, data on POIs visited by users is needed, and POI data is collected only for the driving routes that users set destinations and search for routes.

[0006] However, when using a vehicle, the percentage of customers using route search services is significantly higher than the percentage of customers not using route search services, but when viewed from the perspective of vehicle driving, the percentage of customers driving without a destination is relatively higher than the percentage setting a destination. For example, users will set a destination when driving on a new route, requesting real-time traffic information, or driving a long distance, but they are less likely to set their destination for familiar routes (such as commuter routes that are usually not heavily congested), regular driving routes (such as shops around their home), and short distances.

[0007] For this reason, the percentage of users who use the route search service at least once within a certain period is high, but the percentage of those who set a destination during the total driving time of the vehicle is low, making it difficult to ensure POI data. Summary of the Invention

[0008] The present invention aims to provide an apparatus and method for predicting destinations, which enables the setting of previously visited destinations for each vehicle by predicting and providing destinations without inputting destinations by a connected car service, and also provides secure POI data.

[0009] The purpose of this invention is not limited to the above-described purposes, and other purposes not mentioned will be readily apparent to those skilled in the art from the following description.

[0010] According to one aspect of the invention, an apparatus for predicting a destination is provided, the apparatus comprising: one or more processors; and a memory configured to store one or more programs executed by the one or more processors, wherein the processors can classify trajectory data collected from vehicles using connected car services into multiple driving routes, generate departure information related to a departure point and arrival information related to a destination point for each of the multiple driving routes, and predict one or more driving destinations for each departure information based on at least one of the following: the correlation between the departure information and arrival information generated for each of the multiple driving routes and whether the same driving routes among the multiple driving routes appear periodically.

[0011] The processor can generate departure information based on trajectory data, including at least one of the following features: departure point, departure time interval, and day of the week.

[0012] The processor can generate at least one of the following as arrival information for the first driving route among multiple driving routes: a destination point of interest (POI) set by the vehicle user through the route search service, a past POI with past access history when the route search service is not used, and the destination location of the first driving route.

[0013] When a vehicle user uses the route search service provided by the connected car service to set a destination and then drives along a first driving route, the processor can generate arrival information for the first driving route from the set destination.

[0014] The device may also include a database configured to store a list of POIs that the vehicle has previously visited, wherein when the vehicle is traveling along a first driving route without using the route search service provided by the connected car service, the processor can identify the POI closest to the end point of the first driving route from the POI access list and generate the identified POI as arrival information for the first driving route.

[0015] When a vehicle travels along a first driving route without using the route search service provided by the connected car service, and when there is no POI within a radius of m meters (m is a positive number) from the end point of the first driving route, the processor can generate the end point of the first driving route as the arrival information of the first driving route.

[0016] When the first destination prediction mode is set, the processor can analyze the departure and arrival information generated for each of the multiple driving routes. It calculates the probability of the arrival point by calculating the ratio of the number of times the same first arrival information is generated to the number of times the same first departure information is generated. When the calculated probability of the arrival point is greater than or equal to a first threshold, it determines that the first departure information is related to the first arrival information and predicts the first arrival information as the driving destination of the first departure information.

[0017] When the second destination prediction mode is set, the processor can set the first arrival information as a candidate for the driving destination, and calculate the occurrence interval of the driving route composed of the first departure information and the first arrival information on a daily basis. When the average value of all the calculated occurrence intervals is less than or equal to the second threshold, the driving route composed of the first departure information and the first arrival information is determined as a regular driving route, and finally the first arrival information set as a candidate is predicted as the driving destination of the first departure information.

[0018] The collected trajectory data includes time information indicating the collection time of the trajectory data, and the processor can use the time information indicating the collection time of the first and second trajectory data collected consecutively to distinguish the driving route.

[0019] If a destination is not entered through the connected car service within a certain period after the vehicle engine is started, the processor can determine the current departure information, including the vehicle's current location, current time interval, and current day of the week, based on the current trajectory data collected from the vehicle, and provide the vehicle with a predicted driving destination based on the confirmed current departure information.

[0020] According to one aspect of the invention, a method for predicting a destination is provided, the method being performed by an electronic device including a memory and one or more processors, the memory being configured to store one or more programs executed by the one or more processors, the method comprising: classifying trajectory data collected from vehicles using connected car services into multiple driving routes; generating departure information related to a departure point and arrival information related to a destination point for each of the multiple driving routes; and predicting one or more driving destinations for each departure information based on at least one of the following: the correlation between the departure information and arrival information generated for each of the multiple driving routes and whether the same driving route appears periodically among the multiple driving routes.

[0021] When generating departure and arrival information, departure information can be generated based on trajectory data, including at least one of the following features: departure point, departure time interval, and day of the week.

[0022] When generating departure and arrival information, for the first driving route among multiple driving routes, at least one of the following can be generated as arrival information: a point of interest (POI) set by the vehicle user through the route search service, a past POI with past access history when the route search service is not used, and the destination location of the first driving route.

[0023] When generating departure and arrival information, when a vehicle user uses the route search service provided by the connected car service to set a destination and then drives along a first driving route, the set destination can be generated as arrival information for the first driving route.

[0024] A list of POIs that a vehicle has previously visited can be stored in a database. When generating departure and arrival information, if a vehicle is traveling along a first route without using the route search service provided by the connected car service, the POI closest to the end of the first route can be identified from the POI access list, and the identified POI can be used to generate arrival information for the first route.

[0025] When generating departure and arrival information, if a vehicle is traveling along a first driving route without using the route search service provided by the connected car service, and if there is no POI within a radius of m meters (m is a positive number) from the end point of the first driving route, the end point of the first driving route can be generated as the arrival information of the first driving route.

[0026] When predicting one or more travel destinations, if the first destination prediction mode is set, the departure and arrival information generated for each of the multiple travel routes can be analyzed. The probability of arrival is calculated by calculating the ratio of the number of times the same first arrival information is generated to the number of times the same first departure information is generated. When the calculated probability of arrival is greater than or equal to a first threshold, it can be determined that the first departure information is related to the first arrival information, and the first arrival information can be predicted as the travel destination of the first departure information.

[0027] When predicting one or more travel destinations, if a second destination prediction mode is set, the first arrival information can be set as a candidate travel destination. The occurrence interval of the travel route consisting of the first departure information and the first arrival information can be calculated on a daily basis. When the average of all calculated occurrence intervals is less than or equal to a second threshold, the travel route consisting of the first departure information and the first arrival information can be determined as a regular travel route. Finally, the first arrival information set as a candidate can be predicted as the travel destination of the first departure information.

[0028] The collected trajectory data includes time information indicating the collection time of the trajectory data, and when classifying trajectory data, the time information indicating the collection time of the first and second trajectory data collected consecutively can be used to distinguish the driving routes. Attached Figure Description

[0029] The above and other objects, features and advantages of the present invention will become more apparent to those skilled in the art from the detailed description of exemplary embodiments of the present invention with reference to the accompanying drawings, in which:

[0030] Figure 1 This is a diagram illustrating a configuration in which a device for predicting a destination according to an embodiment of the present invention is connected to enable communication with multiple vehicles;

[0031] Figure 2 This is a block diagram illustrating a device for predicting a destination according to an embodiment of the present invention;

[0032] Figure 3 It is shown in detail Figure 2 The block diagram of the processor is shown below;

[0033] Figure 4 This is a diagram illustrating an example of the probability of arrival of the arrival point for each departure information calculated by the first destination prediction unit;

[0034] Figure 5 This is another example of a diagram illustrating the probability of arrival of the arrival point for each departure information calculated by the first destination prediction unit;

[0035] Figure 6 This is an exemplary diagram showing the occurrence interval of the same driving route, which is composed of the first departure information and the first arrival information, calculated by the second destination prediction unit;

[0036] Figure 7 This is a flowchart illustrating a method for predicting a destination according to an embodiment of the present invention, and illustrates a method for predicting a destination when a first destination prediction mode is set; and

[0037] Figure 8 This is a flowchart illustrating a method for predicting a destination according to another embodiment of the present invention, and also illustrates a method for predicting a destination when a second destination prediction mode is set. Detailed Implementation

[0038] In the following description, embodiments of the invention will be detailed with reference to the accompanying drawings, enabling those skilled in the art to readily practice the invention. However, the invention can be modified in various ways and is not limited to the embodiments described herein.

[0039] Furthermore, in describing exemplary embodiments of the invention, well-known functions or structures will not be described in detail, as they would unnecessarily obscure the understanding of the invention. In the drawings, components irrelevant to the description of the invention are omitted, and similar reference numerals are applied to similar components.

[0040] In this invention, when a component is referred to as "connected," "coupled," or "joined" to another component, this can include not only a direct connection but also an indirect connection where another component is present. Furthermore, when a component "comprises" or "has" another component, this means that the component may also include other components, unless otherwise stated.

[0041] In this invention, terms such as "first" and "second" are used only for the purpose of distinguishing one component from other components, and unless otherwise specified, the order, importance, etc., of the components are not limited. Therefore, within the scope of this invention, a first component in one embodiment may be referred to as a second component in another embodiment, and similarly, a second component in one embodiment may be referred to as a first component in other embodiments.

[0042] In this invention, the components distinguished from each other are intended to clearly explain each feature and do not imply that these components are necessarily separate. That is, multiple components may be integrated to form a single hardware or software unit, or a single component may be distributed to form multiple hardware or software units. Therefore, such integrated or distributed implementations are included within the scope of this invention, even if not described separately.

[0043] In this invention, the components described in the various embodiments are not necessarily essential components, and some components may be optional. Therefore, embodiments consisting of a subset of the components described in the embodiments are also included within the scope of this invention. Furthermore, embodiments including components other than those described in the various embodiments are also included within the scope of this invention.

[0044] In this invention, phrases such as “A or B”, “at least one of A and B”, “at least one of A or B”, “A, B or C”, “at least one of A, B and C”, and “at least one of A, B or C” or combinations thereof may include any one or all possible combinations of the items listed together in the corresponding phrase.

[0045] Referring to the accompanying drawings, various advantages and features of the present invention, as well as methods for implementing them, will become apparent from the following description of embodiments. However, the present invention is not limited to the exemplary embodiments disclosed below, but can be implemented in various different forms. These embodiments are provided only to complete the invention and to allow those skilled in the art to fully understand the scope of the invention.

[0046] Furthermore, in this specification, terms such as “module,” “unit,” and “device” may be intended to refer to the functional and structural combination of hardware and software that drives or is used to drive hardware. For example, hardware in this document may be a data processing device that includes a central processing unit (CPU) or another processor. Additionally, hardware-driven software may refer to a running process, object, executable file, executing thread, program, etc.

[0047] Furthermore, in this specification, unless the context clearly indicates otherwise, the singular form may include the plural form. Similarly, unless the context clearly indicates otherwise, the plural form may include the singular form. Throughout this specification, unless otherwise specified, "including" any component means that other components may also be included, not excluded.

[0048] In this invention, "route search service" refers to a service that provides a route to a destination via navigation provided by a connected car service (CCS). Therefore, "route search service" in this invention does not include services not provided by CCS or provided by separate navigation devices.

[0049] In this invention, "trajectory data" is a set of location information collected by the vehicle-mounted Global Positioning System (GPS) module in units of 1 second. The GPS module collects this information in units of 1 second while the vehicle is in motion.

[0050] In this invention, "navigation data" refers to data generated when a user uses a route search service via navigation linked to the CCS, while providing route guidance to the vehicle.

[0051] In the following text, specific technical details that will be implemented in this invention will be described in detail with reference to the accompanying drawings.

[0052] Figure 1 This is a diagram illustrating a configuration in which a device 100 for predicting a destination according to an embodiment of the present invention is connected to enable communication with a plurality of vehicles 11, 12 and 13.

[0053] refer to Figure 1Multiple vehicles 11, 12 and 13 provide connected car services and can send information and / or data via network 20 to and receive information and / or data from device 100 for predicting destinations.

[0054] Connected cars are vehicles that integrate information and communication technologies to enable two-way communication, mobility services, and more. For example, connected cars can be remotely turned on and off from the outside (by another vehicle or traffic and communication infrastructure), or can control functions such as heating, air conditioning, and navigation, and can provide services such as email, multimedia streaming, and social networking services (SNS).

[0055] For example, multiple vehicles 11, 12, and 13 can send trajectory data related to their driving via network 20 to device 100 for destination prediction. The trajectory data can be collected by vehicles subscribed to CCS, regardless of whether the route search service is used, and is then sent to device 100 for destination prediction. Vehicles begin collecting trajectory data when their engines are on and stop when the vehicles are off. Each trajectory data entry can include information about the time the data was collected and the day of the week.

[0056] Furthermore, when the first vehicle 11 of the multiple vehicles 11, 12, and 13 uses the route search service via navigation linked to the CCS, navigation data, including the user-inputted destination, can be sent to the device 100 for predicting the destination. The navigation data may include the user-inputted destination, information about the time of the input destination, and trajectory data corresponding to the route traveled by the vehicle.

[0057] Network 20 is configured to enable communication between multiple vehicles 11, 12, and 13 and device 100 for predicting destinations, and depending on the installation environment, it can be established using wired networks such as Ethernet, wired home networks (powerline communication), telephone line communication devices, wireless networks such as mobile communication networks (e.g., LTE, 4G, 5G, etc.), wireless LAN (WLAN), Wi-Fi, Wi-Fi Direct, Bluetooth, Ultra Wideband (UWB), Near Field Communication (NFC), Wireless Universal Serial Bus (Wireless USB), or ZigBee, or a combination of wired and wireless networks.

[0058] The device 100 for predicting destinations can use CCS to analyze trajectory and navigation data collected from multiple vehicles 11, 12, and 13 to predict the destination for each departure information. The configuration for predicting the destination for each departure information will be described in detail below. Thereafter, even if the user of the first vehicle 11 providing the CCS does not use the route search service provided by the CCS, i.e., does not input the destination through the CCS, the predicted destination based on the current departure information of the first vehicle 11 can still be provided to the first vehicle 11.

[0059] The first vehicle 11 can encourage the user to use the route search service linked to the CCS by displaying a predicted destination provided by the device 100 for predicting the destination on a display device that provides the CCS. When the user of the first vehicle 11 is not using the route search service provided by the CCS, the user can select the destination displayed on the display device, and the first vehicle 11 can use the route search service provided by the CCS with guidance on the driving route to the destination.

[0060] Figure 2 This is a block diagram illustrating a device 100 for predicting a destination according to an embodiment of the present invention.

[0061] refer to Figure 2 According to an embodiment of the present invention, a device 100 for predicting a destination may include a communication module 110, a database (DB) 120, a memory 130, and a processor 140.

[0062] Communication module 110 can be configured to allow multiple vehicles 11, 12, and 13, as well as a destination prediction device 100, to send and receive information and / or data to and from each other via network 20. Communication module 110 may include wireless communication circuitry and / or wired communication circuitry. For example, communication module 110 can receive multiple trajectory data and multiple navigation data from multiple vehicles 11, 12, and 13. Furthermore, communication module 110 can provide configuration for the destination prediction device 100 to communicate with another system (e.g., a cloud system).

[0063] DB 120 can collect and store multiple trajectory data received from multiple vehicles 11, 12, and 13. The trajectory data can include location information and information about the time and day of the week when the data was collected. Trajectory data can be collected during operation, even when CCS is activated and even when the route search service is not being used.

[0064] In addition, DB 120 can collect and store multiple navigation data streams received from multiple vehicles 11, 12, and 13. The navigation data may include the user-inputted destination, information about the time of the input destination, and trajectory data corresponding to the route traveled by the vehicle.

[0065] In addition, DB 120 can store a list of POI accesses, which lists the points of interest (POIs) that the vehicles 11, 12, and 13 have previously visited. The POI access list can display the name, location information, or address of each POI. Past POIs can be points where users set their destination using a route search service or points set individually by the user.

[0066] The memory 130 may store at least one program (e.g., operating system, software, firmware, middleware, or multiple applications, etc.), various data, and at least one instruction for controlling the device 100 for predicting the destination, and may load programs, read or write data, or perform operations corresponding to instructions upon request from the processor 140. The memory 130 may include volatile memory or non-volatile memory.

[0067] Processor 140 can perform overall control of device 100 based on input commands. Instructions can be provided to processor 140 via memory 130 or communication module 110. For example, processor 140 can execute programs or instructions stored in memory 130 to control the operation of other components (hardware or software) and perform data processing and calculations. Furthermore, processor 140 can load commands or data received from other components 110 to 130 into volatile memory, process the commands or data stored in volatile memory, and store the processing results in non-volatile memory. Processor 140 may include, for example, one or more of at least a CPU, at least a microprocessor, at least a digital signal processor (DSP), at least an application-specific integrated circuit (ASIC), at least a programmable logic device (PLD), and at least a field-programmable gate array (FPGA).

[0068] Processor 140 can use CCS to classify trajectory data collected from multiple vehicles 11, 12, and 13 and stored in DB 120 into multiple driving routes; generate departure and arrival information for each of the multiple driving routes; and predict one or more driving destinations for each departure information based on at least one of the following: the correlation between the departure and arrival information generated for each of the multiple driving routes, and whether the same driving route appears periodically among the multiple driving routes. The departure information is information related to when the vehicle engine is on, and the arrival information is information related to when the vehicle engine is off and the vehicle arrives at its destination.

[0069] Figure 3 It is shown in detail Figure 2 The block diagram of processor 140 shown is shown in the figure.

[0070] refer to Figure 3 The processor 140 may include a trajectory processing unit 141, a departure information generation unit 143, an arrival information generation unit 145, a first destination prediction unit 147, a second destination prediction unit 148, and a destination provisioning unit 149.

[0071] The trajectory processing unit 141 can process or preprocess trajectory data collected from multiple vehicles 11, 12, and 13 and stored in the DB 120 to generate multiple driving routes. The trajectory processing unit 141 can use the time information of the first and second trajectory data collected consecutively from the trajectory data stored in the DB 120 to distinguish the driving routes.

[0072] For example, when the difference between the time information of the continuously collected first trajectory data and the second trajectory data exceeds a preset reference value, the trajectory processing unit 141 determines that the first trajectory data and the second trajectory data correspond to different driving routes. Therefore, the trajectory processing unit 141 can identify previously collected trajectory data including the first trajectory data as one driving route, and identify later collected trajectory data including the second trajectory data as another driving route. The preset reference value can be, for example, 15 minutes, which means that no trajectory data is collected within 15 minutes after the vehicle engine has been turned off. 15 minutes is an example and is not limited to it, and can be changed.

[0073] On the other hand, if the difference between the time information of the collected first trajectory data and the second trajectory data is less than or equal to the reference value, the trajectory processing unit 141 can determine that the first trajectory data and the second trajectory data correspond to the same driving route. In other words, the trajectory processing unit 141 can process it as if the vehicle has stopped for a period of time and then continues to drive towards the same destination.

[0074] The departure information generation unit 143 can generate departure information for each of the multiple driving routes generated by the trajectory processing unit 141. The departure information generation unit 143 can generate departure information based on the trajectory data of the driving routes and store the departure information in the DB 120. The departure information may include at least one of the following features: departure point, departure time interval, and day of the week.

[0075] For example, the departure information generation unit 143 can generate the first trajectory data (i.e., the address or administrative area of ​​the first GPS coordinate) from the trajectory data corresponding to the first driving route among multiple driving routes as the departure point. Furthermore, the departure information generation unit 143 can confirm the time information of the occurrence of the first GPS coordinate and generate the time interval to which the confirmed time information belongs as the departure time interval. The time interval is divided into, for example, dawn (01:00-05:00), morning (05:00-09:00), AM (09:00-12:00), afternoon (12:00-17:00), evening (17:00-20:00), and night (20:00-1:00 the next day). In addition, the departure information generation unit 143 can confirm the day of the week in which the first GPS coordinate occurred and further confirm the day-of-week characteristics to which the confirmed day of the week belongs. The day-of-week characteristics can be divided into, for example, weekdays, weekends, and holidays. Therefore, when trajectory data corresponding to the first driving route is collected at 10:00 AM on Monday, August 1st at AA Cave (administrative cave name), the departure information generation unit 143 can generate departure information consisting of AA Cave, AM, and weekday.

[0076] Arrival information generation unit 145 can generate arrival information for each of the multiple driving routes. Arrival information generation unit 145 can generate arrival information for each driving route based on trajectory data and navigation data corresponding to the driving route, and store the generated arrival information in DB 120.

[0077] Arrival information may include the destination POI set by the vehicle user through the route search service when traveling along the first of multiple routes, past POIs with past access history when the route search service is not used, one of the destination locations of the first route, and the arrival time interval.

[0078] Specifically, when a vehicle user sets a destination using the route search service provided by CCS and then drives along a first driving route, the arrival information generation unit 145 can generate arrival information for the first driving route from the set destination (i.e., destination POI) and the arrival time interval of the destination POI. The arrival information generation unit 145 can confirm, based on navigation data, whether a destination has been set by the vehicle user using the route search service. The arrival information generation unit 145 can search the DB 120 for navigation data matching the departure information of the first driving route, and when the searched navigation data includes a destination, it can determine that the user has set a destination and is driving along the first driving route.

[0079] Alternatively, when a vehicle user travels along a first driving route without using the route search service, the arrival information generation unit 145 can identify the POI closest to the end of the first driving route within a radius of m meters (m is a positive number) based on the end location of the first driving route in the POI access list, and generate the identified POI (i.e., past POI) as arrival information for the first driving route. The end location of the first driving route can be obtained from the last collected trajectory data in the trajectory data of the first driving route.

[0080] Alternatively, when a vehicle user is traveling along a first driving route without using the route search service and when there is no POI located within a radius of m meters from the end point of the first driving route, the arrival information generation unit 145 may generate the end point location of the first driving route (e.g., the address of the end point) as arrival information for the first driving route.

[0081] When the first destination prediction mode is set, the first destination prediction unit 147 can analyze the departure and arrival information generated for each of the multiple driving routes to generate multiple different arrival information for the same departure information. Furthermore, it can calculate the probability of arrival by calculating the ratio of the number of times the same arrival information is generated to the number of times the same departure information is generated, thereby calculating the probability of arrival for all arrival information. The first destination prediction mode can be set individually for multiple vehicles 11, 12, and 13 or collectively for the device 100 used for destination prediction.

[0082] For example, the first destination prediction unit 147 analyzes the departure and arrival information generated for each of the multiple driving routes, and calculates the probability of arrival by calculating the ratio of the number of times the same arrival information (hereinafter referred to as "first arrival information") is generated to the number of times the same departure information (hereinafter referred to as "first departure information") is generated. The first departure information is one of the multiple departure information generated based on the multiple driving routes, and the first arrival information is one of the multiple arrival information generated based on the multiple driving routes, and the first departure information and the first arrival information correspond to the same driving route.

[0083] When the calculated probability of arrival is greater than or equal to a first threshold, the first destination prediction unit 147 determines that the first departure information and the first arrival information are related, predicts the first arrival information as the destination of the first departure information, and then maps the predicted first arrival information to the first departure information and stores the predicted first arrival information in DB 120. That is, when the vehicle starts driving under the rule corresponding to the first departure information, the first destination prediction unit 147 can predict that the vehicle's destination will be the first arrival information (i.e., the destination POI, past POI, or end point location included in the first arrival information).

[0084] Furthermore, when the calculated probability of the arrival point is less than the first threshold, the first destination prediction unit 147 may not predict the destination of the first departure information.

[0085] Figure 4 This is a diagram illustrating an example of the probability of arrival of each departure information calculated by the first destination prediction unit 147.

[0086] Figure 4 The diagram illustrates the collection of N days' worth of trajectory data (N being a positive number greater than or equal to 1) from vehicles using the route search service provided by CCS, and the calculation of the probability of arrival points by utilizing the collected trajectory data (i.e., departure and arrival information for multiple routes). In this case, arrival information may include destination POIs.

[0087] refer to Figure 4 The vehicle index is an index of multiple vehicles (11, 12, and 13) that have subscribed to CCS and have a history of using the route search service, and can be represented by multiple vehicle IDs. The departure rule is the departure information generated for each route, including the departure point, time interval, and day of the week. The arrival point is the location or address included in the arrival information generated for each route.

[0088] The probability of reaching the destination is the percentage of times the same arrival information is generated across all driving routes, compared to the percentage of times the same departure rule is generated. For example, if the number of times a vehicle departs from Negok-dong on a weekday afternoon is 80, the number of times it arrives at Bae-dong is 64 out of 80, and therefore the probability of reaching the destination is 64 / 80, or 80%. Furthermore, since the number of times a vehicle travels from Negok-dong to Myeong-dong on a weekday afternoon is 16 out of 80, the probability of reaching the destination is 16 / 80, or 20%. A 20% probability of reaching the destination means that if the engine is started in Negok-dong on a weekday afternoon, there is a 20% probability that the vehicle will arrive in Myeong-dong (i.e., the destination).

[0089] The first destination prediction unit 147 can adjust the first threshold based on the average of the arrival point occurrence probabilities calculated according to the frequency of departure rules and arrival information. For example, the first destination prediction unit 147 can set the rounded-up value of the arrival point occurrence probability as the first threshold. Figure 4 In the case where the average probability of arrival at the destination is 55.35%, the first destination prediction unit 147 may set the first threshold to 60%. Alternatively, the first destination prediction unit 147 may of course use a first threshold that is set without considering the average probability of arrival at the destination.

[0090] When the first threshold is 60%, if the calculated probability of the arrival point is greater than 60%, the first destination prediction unit 147 can determine that the first departure information is related to the first arrival information, and predict the first arrival information as the destination of the first departure information. Figure 4 In this case, for vehicles departing from Naegok-dong on a weekday afternoon, Bae-dong can be predicted as the destination.

[0091] Figure 5 This is a diagram illustrating another example of the probability of arrival of arrival information for each departure information calculated by the first destination prediction unit 147.

[0092] Figure 5 The diagram illustrates the collection of N days of trajectory data from vehicles using CCS but not the route search service, and the calculation of the probability of arrival points by utilizing the collected trajectory data (i.e., departure and arrival information for multiple routes). In this case, arrival information may include either a past POI or the destination location of the route.

[0093] like Figure 5 As shown, the first destination prediction unit 147 can calculate the probability of arrival of a destination for trajectory data (i.e., driving routes) collected from vehicles that have never used the route search service, and predict the driving destination by comparing the calculated probability of arrival of a destination with a first threshold.

[0094] Simultaneously, when the second destination prediction mode is set, the first destination prediction unit 147 can select departure information (i.e., departure rules) and arrival information (i.e., arrival points) whose probability of occurrence is greater than or equal to a first threshold, and send the selected departure information and arrival information to the second destination prediction unit 148. The second destination prediction mode can be set individually in multiple vehicles 11, 12, and 13 or jointly in the device 100 used for predicting destinations.

[0095] The second destination prediction unit 148 can set the first arrival information predicted by the first destination prediction unit 147 as a candidate for the driving destination, and calculate the occurrence interval of the driving route composed of the first departure information and the first arrival information on a daily basis. When the average of all calculated occurrence intervals is less than or equal to a second threshold, the second destination prediction unit 148 can determine the driving route composed of the first departure information and the first arrival information as a regular driving route, and finally predict the first arrival information set as a candidate as the driving destination of the first departure information.

[0096] Figure 6 This is an exemplary diagram showing the occurrence interval of the same driving route, which is composed of the first departure information and the first arrival information, calculated by the second destination prediction unit 148.

[0097] refer to Figure 6 The vehicle index is an index of vehicles that generate an arrival probability greater than or equal to a first threshold. The request date, day of the week, and request time are the date, day of the week, and request time when the same route, consisting of the first departure information and the first arrival information, occurs. The occurrence interval is the date difference when multiple routes, including the first departure information and the first arrival information, occur. Figure 6 In this context, the intervals of occurrence are 2 days, 1 day, 4 days, and 1 day. The departure point is the departure point included in the first departure information, and the destination can be the arrival point included in the first arrival information (or the destination predicted by the first destination prediction unit 147). The time interval is the time interval in which the first departure information occurs, and all time intervals are the same as morning, and the day of the week feature is the day of the week feature in which the first departure information occurs, and all day of the week features are the same as weekdays.

[0098] When the second threshold is, for example, 3 days, if the average of the calculated occurrence intervals is less than or equal to 3 days, the second destination prediction unit 148 can determine that the travel route consisting of the first departure information and the first arrival information occurs regularly every working day. Therefore, the second destination prediction unit 148 can predict the first arrival information, which is set as a candidate, as the final destination of the first departure information, map the first departure information and the first arrival information, and store the first departure information and the first arrival information in DB 120.

[0099] When the vehicle engine starts at the starting point of the first departure information, the first departure information and first arrival information (i.e., the final destination) mapped and stored in DB 120 can be provided to the vehicle as the driving destination. Specifically, when no destination is entered through the route search service after the vehicle engine is started, the destination providing unit 149 can confirm the current departure information (i.e., departure rule) including the vehicle's current location information, current time interval, and current day of the week based on the current trajectory data collected from the vehicle, and confirm the mapped and stored driving destination of the same departure information as the confirmed current departure information in DB 120, and provide the mapped and stored driving destination to the vehicle.

[0100] When the destination displayed on the vehicle matches the user's intended destination, the user can click on the destination, and the vehicle can activate the route search service and begin route guidance.

[0101] Figure 7 This is a flowchart illustrating a method for predicting a destination according to an embodiment of the present invention, and also illustrates a method for predicting a destination when setting a first destination prediction mode.

[0102] Execution in Figure 7 The electronic device for predicting the destination shown in the figure can be a reference. Figures 1 to 6 The device 100 described is used to predict the destination.

[0103] refer to Figure 7 Electronic devices can use CCS to collect trajectory data from multiple vehicles (S705).

[0104] The electronic device can analyze the collected trajectory data to generate multiple driving routes (S710).

[0105] The electronic device can generate departure information (S715) that includes information related to the departure of each of the multiple driving routes. In operation S715, departure information can be generated based on trajectory data, including at least one of the following characteristics: the vehicle's departure point, the departure time interval, and the day of the week of departure.

[0106] When a vehicle user uses the route search service provided by CCS to set a destination and then drives along a route (e.g., the first route among multiple routes) (S720 - Yes), the electronic device can determine the set destination as the destination POI (S725), and then use the destination POI to generate arrival information for the first route (S730). Operation S730 may include the destination POI and the time interval for arriving at the destination POI.

[0107] When no destination is set in operation S720, that is, when the vehicle is traveling along the first driving route without using the route search service provided by CCS (S720 - No), if there is a POI within a radius of m meters (where m is a positive number) based on the end point of the first driving route (S735 - Yes), the electronic device can confirm the POI closest to the end point of the first driving route from the POI access list and generate the confirmed POI as arrival information for the first driving route (S740 and S730).

[0108] Furthermore, when the vehicle is traveling along the first driving route without using the route search service provided by CCS, if there is no POI within a radius of m meters in operation S735 (S735-No), the electronic device can generate the destination location of the first driving route as arrival information for the first driving route (S745 and S730).

[0109] When arrival information is generated for all driving routes by repeatedly executing operations S720 to S745, the electronic device can predict one or more driving destinations for each departure information based on the correlation between departure information and arrival information generated for each of the multiple driving routes (S750 to S760).

[0110] That is, when the first destination prediction mode is set, the electronic device can analyze the departure and arrival information generated for each of the multiple driving routes, and calculate the probability of the arrival point appearing by calculating the ratio of the number of times the same first arrival information is generated to the number of times the same first departure information is generated (S750). The first departure information is one of the multiple departure information generated based on the multiple driving routes, and the first arrival information is one of the multiple arrival information generated based on the multiple driving routes, and the first departure information and the first arrival information correspond to the same driving route.

[0111] When the probability of arrival calculated in operation S750 is greater than or equal to the first threshold (S755-Yes), the electronic device determines that the first departure information is related to the first arrival information.

[0112] Furthermore, the first arrival information can be predicted as the destination of the first departure information (S760).

[0113] On the other hand, when the probability of the arrival point calculated in operation S750 is less than the first threshold (S755-No), the electronic device determines that the first departure information and the first arrival information are unrelated or have low correlation, and can end the operation of predicting the destination associated with the first departure information (S765).

[0114] Figure 8This is a flowchart illustrating a method for predicting a destination according to another embodiment of the present invention, and also illustrates a method for predicting a destination when a second destination prediction mode is set.

[0115] Execution in Figure 8 The electronic device for predicting the destination shown in the figure can be a reference. Figures 1 to 6 The device 100 described is used to predict the destination.

[0116] refer to Figure 8 When the second destination prediction mode is set, the electronic device can predict one or more destinations for each departure information based on at least one of the following: the correlation between departure and arrival information generated for each of multiple routes, and whether the same route appears periodically among the multiple routes. Therefore, the electronic device can perform reference... Figure 7 The operations described in steps S705 to S755 are executed afterward. Figure 8 The operation shown in the figure.

[0117] Because a second destination prediction mode is set, the electronic device can set the driving destination predicted in operation S760 as a candidate (S810).

[0118] The electronic device can calculate the occurrence interval of the same travel route consisting of the first departure information and the first arrival information on a daily basis (S820), and calculate the average value of all the calculated occurrence intervals (S830).

[0119] When the average value of the generated occurrence intervals is less than or equal to the second threshold (S840 - Yes), the electronic device can determine that the driving route consisting of the first departure information and the first arrival information is a regular driving route (S850), and finally predict the first arrival information set as a candidate as the driving destination of the first departure information in operation S810 (S860).

[0120] On the other hand, when the average value of the generated occurrence intervals exceeds the second threshold (S840-No), the electronic device can terminate the operation of predicting the destination associated with the first departure information, because the driving route composed of the first departure information and the first arrival information is not a driving scenario that occurs regularly (S870).

[0121] According to the present invention, even without inputting a destination through the connected car service, previously visited destinations can be predicted and provided for each vehicle, and users are encouraged to set their destinations in the route search service provided by the connected car service to ensure POI data.

[0122] In addition, according to the present invention, the percentage of driving without setting a destination on frequently visited or regularly driven routes while using route search services can be reduced, and the destination POI can be ensured every time the vehicle is driven and the destination POI can be used to understand driving behavior.

[0123] Furthermore, according to the present invention, by taking into account departure rules (i.e., departure point, departure time interval, and departure date) to predict the destination and then guiding the user to the predicted destination, the destination can be predicted more accurately and precisely, and thus, user convenience is improved.

[0124] Furthermore, according to the present invention, the use of navigation linked to connected car services can be promoted by suggesting the predicted destination to customers who do not use navigation linked to connected car services.

[0125] The effects achievable by this invention are not limited to those described above. That is, through the following description, those skilled in the art will clearly understand other objectives not described.

[0126] For clarity of explanation, the exemplary method of the present invention described above is expressed as a series of operations; however, this is not intended to limit the order of the steps, and the steps may be performed simultaneously or in different orders if necessary. To implement the method according to the invention, in addition to the steps illustrated, other steps may be included, some steps may be omitted and others included, or some steps may be omitted and other additional steps included.

[0127] The various embodiments of the present invention are intended to explain representative aspects of the invention, rather than to list all possible combinations, and the contents described in the different embodiments may be applied independently or in combination of two or more.

[0128] Furthermore, various embodiments of the present invention can be implemented using hardware, firmware, software, combinations thereof, etc. For hardware implementation, various embodiments of the present invention can be implemented using one or more ASICs, DSPs, digital signal processing devices (DSPDs), PLDs, FPGAs, processors, controllers, microcontrollers, microprocessors, etc.

[0129] The scope of this invention includes software or machine-executable instructions (e.g., operating systems, applications, firmware, programs, etc.) that enable the operation of methods according to various embodiments to be performed on a device or computer, as well as non-transitory computer-readable media in which such software, instructions, etc. are stored and can be executed on a device or computer.

Claims

1. A device for predicting a destination, comprising: One or more processors; as well as The memory is configured to store one or more programs executed by the one or more processors. The processor is configured as follows: The trajectory data collected from vehicles using connected car services will be categorized into multiple driving routes. For each of the multiple driving routes, generate departure information related to the starting point and arrival information related to the destination. Based on at least one of the following: the correlation between the departure information and the arrival information generated for each of the multiple driving routes, and whether the same driving route among the multiple driving routes appears periodically, one or more driving destinations for each departure information are predicted. and Based on whether a destination has been entered, it is determined whether to automatically provide a predicted driving destination to control the operation of the vehicle.

2. The device according to claim 1, wherein, Based on the trajectory data, the processor generates departure information including at least one of the following features: departure point, departure time interval, and day of the week.

3. The device according to claim 1, wherein, For a first driving route among the plurality of driving routes, the processor generates at least one of the following as arrival information: a destination point of interest set by the user of the vehicle through a route search service, a past point of interest with past access history when the route search service is not used, and the destination location of the first driving route.

4. The device according to claim 3, wherein, Based on the user of the vehicle using the route search service provided by the connected car service to set a destination and then drive along the first driving route, the processor generates the set destination as the arrival information for the first driving route.

5. The device according to claim 3, further comprising: The database is configured to store a list of points of interest (POIs) accessed, which lists the POIs that the vehicle has previously visited. Wherein, based on the fact that the vehicle is traveling along the first driving route without using the route search service provided by the connected car service, the processor identifies the point of interest closest to the destination of the first driving route from the point of interest access list, and generates the identified point of interest as the arrival information of the first driving route.

6. The device according to claim 5, wherein, Based on the fact that the vehicle is traveling along the first driving route without using the route search service provided by the connected car service, and based on the fact that there are no points of interest within a radius of m meters from the end point of the first driving route, the processor generates the end point of the first driving route as the arrival information of the first driving route, where m is a positive number.

7. The device according to claim 1, wherein, Based on the setting of the first destination prediction mode. The processor analyzes the departure and arrival information generated for each of the multiple driving routes, and calculates the probability of arrival by measuring the ratio of the number of times the same first arrival information is generated to the number of times the same first departure information is generated. Based on the calculated probability of the arrival point being greater than or equal to a first threshold, it is determined that the first departure information is related to the first arrival information, and the first arrival information is predicted as the driving destination of the first departure information.

8. The device according to claim 7, wherein, Based on the setting of the second destination prediction mode. The processor sets the first arrival information as a candidate for the driving destination, and The occurrence interval of the travel route composed of the first departure information and the first arrival information is calculated on a daily basis, and Based on the fact that the average of all calculated occurrence intervals is less than or equal to a second threshold, the driving route consisting of the first departure information and the first arrival information is determined as a regular driving route, and the first arrival information set as the candidate is finally predicted as the driving destination of the first departure information.

9. The device according to claim 1, wherein, The collected trajectory data includes time information indicating the collection time of the trajectory data, and The processor uses time information indicating the collection time of the first and second trajectory data collected consecutively to distinguish the driving routes.

10. The device according to claim 1, wherein, If the destination is not entered through the connected car service within a specific time after the vehicle engine is started, the processor confirms the current departure information, including the vehicle's current location, current time interval, and current day of the week, based on the current trajectory data collected from the vehicle, and provides the vehicle with a predicted driving destination based on the confirmed current departure information.

11. A method for predicting a destination, the method being performed by an electronic device including a memory and one or more processors, the memory being configured to store one or more programs executed by the one or more processors, the method comprising: The trajectory data collected from vehicles using connected car services will be categorized into multiple driving routes; For each of the multiple driving routes, generate departure information related to the starting point and arrival information related to the destination point; Based on at least one of the following: the correlation between the departure information and the arrival information generated for each of the multiple driving routes, and whether the same driving route among the multiple driving routes appears periodically, one or more driving destinations for each departure information are predicted. as well as Based on whether a destination has been entered, it is determined whether to automatically provide a predicted driving destination to control the operation of the vehicle.

12. The method according to claim 11, wherein, When generating the departure information and the arrival information, the departure information is generated based on the trajectory data, including at least one of the following features: departure point, departure time interval, and day of the week.

13. The method according to claim 11, wherein, When generating the departure information and the arrival information, for the first driving route among the multiple driving routes, at least one of the following is generated as the arrival information: a destination point of interest set by the user of the vehicle through the route search service, a past point of interest with past access history when the route search service is not used, and the destination location of the first driving route.

14. The method according to claim 13, wherein, When generating the departure information and the arrival information, the user of the vehicle uses the route search service provided by the connected car service to set a destination and then drives along the first driving route, generating the set destination as the arrival information for the first driving route.

15. The method according to claim 13, wherein, A list of points of interest accessed by the vehicle in the past is stored in a database, and When generating the departure information and the arrival information, based on the fact that the vehicle travels along the first driving route without using the route search service provided by the connected car service, the point of interest closest to the destination of the first driving route is identified from the point of interest access list, and the identified point of interest is generated as the arrival information of the first driving route.

16. The method according to claim 15, wherein, When generating the departure information and the arrival information, based on the fact that the vehicle travels along the first driving route without using the route search service provided by the connected car service, and based on the fact that there are no points of interest within a radius of m meters from the end point of the first driving route, the end point of the first driving route is generated as the arrival information of the first driving route, where m is a positive number.

17. The method according to claim 11, wherein, When predicting the one or more driving destinations, a first destination prediction mode is set. The probability of arrival is calculated by analyzing the departure and arrival information generated for each of the multiple driving routes, and by calculating the ratio of the number of times the same first arrival information is generated to the number of times the same first departure information is generated. Based on the calculated probability of the arrival point being greater than or equal to a first threshold, it is determined that the first departure information is related to the first arrival information, and the first arrival information is predicted as the destination of the first departure information.

18. The method according to claim 17, wherein, When predicting the one or more driving destinations, a second destination prediction mode is set. Set the first arrival information as a candidate for the driving destination. The occurrence interval of the travel route composed of the first departure information and the first arrival information is calculated on a daily basis, and Based on the fact that the average of all calculated occurrence intervals is less than or equal to a second threshold, the driving route consisting of the first departure information and the first arrival information is determined as a regular driving route, and the first arrival information set as the candidate is finally predicted as the driving destination of the first departure information.

19. The method according to claim 11, wherein, The collected trajectory data includes time information indicating the collection time of the trajectory data, and When classifying the trajectory data, time information indicating the collection time of the first and second trajectory data collected consecutively is used to distinguish the driving routes.

20. The method according to claim 11, wherein, Based on the fact that the destination was not input through the connected car service within a specific time after the vehicle engine was started, the current departure information, including the vehicle's current location information, current time interval, and current day of the week, was confirmed based on the current trajectory data collected from the vehicle, and the predicted driving destination was provided to the vehicle based on the confirmed current departure information.

Images