Method for route planning and navigation device

Abstract

To provide a navigation route planning method and system.SOLUTION: The method includes the steps of: dividing each of a plurality of navigation recommendation routes into at least one road section; and determining a navigation travel route from among the plurality of navigation recommendation routes on the basis of past driving information of a user and label information added by the user for the at least one road section.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to intelligent driving technology, and in particular to navigation route planning. [Background technology]

[0002] Currently, navigation systems provide great convenience for users in planning routes when going out, improving travel efficiency and avoiding traffic congestion and traffic accidents. In conventional navigation route planning, a user typically sets a start point and an end point, and a navigation algorithm automatically plans several candidate recommended routes, allowing the user to select one and then begin navigation. However, in some cases, the candidate routes planned by the conventional algorithm may not be in line with the user's actual needs or driving habits. Summary of the Invention [Means for solving the problem]

[0003] The present invention provides a navigation system that can generate recommended routes that better suit a user's actual needs and driving habits based on accumulated user driving route data, i.e., past driving information and past driving sensations. In this way, by generating recommended routes for a user in situations where a certain amount of user data has been accumulated, it is possible to avoid inaccurate planning times and poor driving experiences due to special road conditions or special events. Furthermore, in areas where a user frequently drives, such as commuter roads, it is advantageous for the user to search for different routes, and labeling the user's route driving sensations satisfies the user's personal needs. [Brief explanation of the drawings]

[0004] [Figure 1A] Schematically shows multiple routes between a start point and an end point. [Figure 1B]Schematically shows multiple routes between a start point and an end point. [Figure 2] 1 shows a schematic diagram of a route planning flowchart. [Figure 3] 10 shows a schematic flow chart of a navigation driving route selection according to an example; [Figure 4] 1 is a block diagram of a navigation system; DETAILED DESCRIPTION OF THE INVENTION

[0005] In conventional navigation systems, the navigation system calculates at least one recommended route based on objective factors that affect driving time, such as road construction status, weather conditions, traffic accident rates, road section speed limits, speed measurement locations for road sections, and road levels, in response to start and end points set by the user, and estimates the expected driving time (ET) of the route. When the user decides or selects one of the recommended routes as the navigation driving route, the navigation system presents the expected driving time (ET) of that driving route, continuously updates the remaining arrival time during the driving, and presents the actual driving time (AT) after the journey is completed. However, conventional navigation systems do not record or store information about the length of the user's previous driving routes as the user's past data, nor do they separately record and analyze each road section in the route.

[0006] Meanwhile, the navigation system proposed by the present invention is provided with a function for recording user driving information, which includes, as past data, time length information (including the expected driving time ET and the actual driving time AT) for each road section of the entire driving route, and further provides the user with the ability to add label information to the road sections. Thus, when the user uses the navigation service, the navigation system generates an optimal recommended route that meets the user's wishes according to the user's past data and the user's label information. According to one example of the present invention, the navigation system can create a history database (DB) for the current vehicle or user in a local or remote server of the in-vehicle infotainment system to store user-specific information, including the user's driving history data and road section label information.

[0007] The recording of time length information and label information for each road section by the navigation system of the present invention will be explained below using as an example the route from point A to point E that is frequently used by users when commuting, as shown in Figure 1A. As shown in the figure, R1: A → B → D → E and R2: A → B → C → E are shown as two frequently used routes from point A to point E.

[0008] According to one example of the present invention, the navigation system divides each route R into multiple road segments according to branching points along the travel route, where a branching point may be a branching intersection with a traffic light or any node that may lead to a new route. For example, the navigation system divides route R1 into two road segments: (1) A→B and (2) B→D→E, and divides route R2 into two road segments: (1) A→B and (2) B→C→E. It should be noted that any suitable rule can be adopted to divide the route, such as a combination rule of branching points and a maximum / minimum distance specification, where the specification value may be, for example, kilometers or time. According to the combination rule, if the branching road segment is smaller than the minimum specification value, the branching point is ignored; if the branching road segment is larger than the specification value, the road segment is further divided, i.e., a branching point is added (e.g., road segment B→C→E is further divided into two road segments: B→C and C→E), except when the road segment is not branching (e.g., a closed road segment).

[0009] When the user actually drives these road segments, the navigation system records data about each road segment. For example, if the user drives two routes, R2 (ABCE) and R1 (ABDE), on August 17th and 18th, the navigation system records the following information in the database DB, as shown in Table I:

[0010] [Table 1]

[0011] According to one embodiment of the present invention, when the navigation system determines that there is a large difference between the actual driving time AT and the expected driving time ET of a road section, it activates a user label mode, e.g., by opening a new user interface (or window) to allow the user to label the current abnormal road section. The content of the label can be determined by the user based on their own observations and experiences and actively input into the navigation system. For example, for the road section AB driven at 2:00 PM on August 17, the user can simply label it as "Large Event in Progress" in the label mode, clearly explaining why the road section AB took longer than expected. Of course, the user's annotation of road sections here is not limited to the difference in time length. The user can also activate and annotate a road section in the navigation route. For example, for the road section BCE, which is of poor overall quality and has difficult sections such as narrow roads, the user can label the road section as "Narrow Roads" even though the time it takes to travel through the road section BCE does not exceed the expected time. Thus, still taking the route from point A to point E as an example, the navigation system can record the corresponding label for each road segment in the database DB, as shown in Table II.

[0012] [Table 2]

[0013] It should be noted that, according to a different embodiment of the present invention, with the above-mentioned past driving data and the accompanying label information, the navigation system can prompt the user to pay attention to each abnormal road segment at the end of the entire journey, for example, by presenting all road segments or abnormal road segments on the graphical user interface GUI of the navigation system and allowing the user to label the abnormal road segments or interesting road segments, so that the navigation system can record the past data and possible labels for each road segment during the user's usual driving, and plan the route for the user's next outing.

[0014] It should also be noted that in the present invention, labeling of road segments is not limited to the above exemplary active user input, and label generation methods may also include, but are not limited to, input based on recognition and understanding of voice, image, video, physical state, and other sensors inside and outside the vehicle. For example, when utilizing conventional or future artificial intelligence (AI) technology, the navigation system can collect information such as the user's conversation and facial expressions while traveling along a road segment, recognize it using a pre-trained machine learning model, and automatically label the corresponding road segment.

[0015] According to a further embodiment of the present invention, as shown in Table II, the labels customized by the user and added to each road section are divided into two types: (1) objective labels T, which represent objective factors that affect the travel time, and (2) subjective labels S, which represent subjective factors that affect the travel time. Here, the objective factors indicated by the objective labels T may be the duration of the red light, the number of intersections, the number of vehicles cutting in, etc. For example, in addition to "long red light" shown in the table, the objective labels T may specifically be "many intersections" or "many cutting in." Furthermore, the subjective factors indicated by the subjective labels S may be "no separation between pedestrians and vehicles," in addition to "large-scale event in progress" and "narrow roads present." It should be noted that the objective labels T and subjective labels S can be classified based on any rules customized by the user. For example, such a rule could be "whether it will cause a long stop." Therefore, a label such as "long red light" would be classified as the objective label T because it would necessarily cause a long stop and wait for the light to change. Labels such as "narrow road" or "no separation between pedestrians and vehicles" would be classified as the subjective label S because they may not cause a long stop and the vehicle may continue traveling (although perhaps not at a high speed) without needing to stop. Furthermore, labels added by users can also be automatically classified using machine learning based on artificial intelligence. Furthermore, as can be seen from Table II, even for the same road section, e.g., AB, users may assign different labels, or even different types of labels, at different travel times.

[0016] According to an embodiment of the present invention, multiple levels of tolerance can be further set for the subjective perception label S. For example, levels 1 to 10 can be set, with 1 being completely acceptable and 10 being completely unacceptable. While the levels 1 to 10 are set for ease of calculation, they can also be optimized according to actual conditions. This allows specific labels belonging to the subjective perception label S of each road section to be assigned a corresponding degree level or degree value. For example, if "large-scale event is being held," the degree value "1" is assigned, indicating that the event is merely a random occurrence and therefore completely acceptable for the road section A and B on the current date. On the other hand, a high degree value "9" is assigned to the user's label "narrow roads present," indicating that the user does not want to travel there. In this way, for a specific scenario in which past user activity data has been recorded, the optimal driving route for the user can be determined based on this past information, label classification, and label evaluation information.

[0017] FIG. 2 shows a flowchart of a route planning method executed by a navigation system according to an example of the present invention. As shown, in step 201, user input information is received, where the input information includes travel destination information and, optionally, may further include other necessary information, such as information about the user's intermediate stops. Typically, the navigation system automatically sets the user's current location as the starting point by default, but if the user wants to set a different starting point, the user can input this into the navigation system. Here, the navigation system can receive the user's input in various forms, such as text or voice.

[0018] In step 202, a plurality of recommended navigation routes R1 to R2 are selected based on the end point information, start point information, and waypoints (if any) input by the user. m Here, m routes are generated, and m routes are shown. Here, a conventional algorithm is used to generate m recommended routes R1 to R2. mStill, let's take the example of a user wanting to start from point A and go to point E, as shown in Figure 1B, and the navigation system will recommend the following three routes based on the traditional navigation algorithm: R1: A → B → D → E R2: A → B → C → E R3: A → F → D → E

[0019] Next, in step 203, each of the three recommended routes R1 to R3 can be divided into a plurality of road sections using preset rules, and as described above, route R1 is divided into road sections A → B and B → D → E, route R2 is divided into road sections A → B and B → C → E, and route R3 is further divided into road sections A → F and F → D → E. Next, the process proceeds to step 204, where the preferred road sections selected by the user are received.

[0020] According to one example of the present invention, in step 204, first, past driving information for road sections R1 to R3 and label information previously added by the user are presented on the graphical interface (GUI) of the navigation system. For example, in this example, the contents of Table III below can be presented.

[0021] [Table 3]

[0022] Based on the road section information for each of the three recommended routes presented on the GUI, the user can tap on the GUI screen to select a road section based on its actual travel time and associated labels (e.g., tap the screen in the GUI to select a road section). For example, the user may determine that road section AB is preferable based on the annotations "large-scale event in progress" and "long red light periods" attached to road section AB on August 17 and 18, as well as the actual travel time of road section AB on August 18. Road section AF also has an actual travel time of 10 minutes, but the user may not select it due to the associated labels reflecting the road's travel experience because the road has a poor driving experience due to the "large number of electric vehicles" on the road. Accordingly, the user intuitively understands that road section BDE has a better actual travel time than the expected time, and road section BCE not only has a longer travel time but also has narrow roads, so the user selects route BDE as the next road section.

[0023] After the user selects all road segments in the GUI, in step 205, the navigation system generates the final driving route A→B→D→E and presents it on the navigation system's graphical interface. After the user accepts this and issues a navigation command, the navigation system begins operation. According to the present invention, even if route R3 takes less time than route R1 (A→B→D→E) based on historical data, the user may be hesitant to select it, but a conventional navigation system may offer R3 to the user as a preferred or default route.

[0024] In the above example, road segment information is presented to the user in an intuitive manner, making it easier for the user to select road segments based on road labels that reflect their past driving experiences and ultimately form a navigation route. In another example of the present invention, a navigation system can automatically determine the optimal navigation route for the user by digitizing such road segment labels. FIG. 3 shows a flowchart of a navigation route planning process according to this implementation, which will still be described with reference to FIG. 1B. Similar to steps 201, 202, and 203, step 301 first receives user input information, including information such as the driving destination, intermediate stops, and starting point, and generates multiple feasible recommended navigation routes, e.g., three routes R1-R3, based on a conventional navigation algorithm. Next, each of the three recommended routes R1-R3, route R, is divided into multiple road segments. For example, R1 is divided into u road segments, R2 is divided into v road segments, and R3 is divided into w road segments.

[0025] In step 303, for one of the three recommended routes, e.g., R1, the objective travel time (herein also referred to as "objective travel time length") Yi (where i=1, 2, ... u) of each road section i of the u road sections included is calculated, where, according to this example of the present invention, the objective travel time Yi is calculated by taking into account a combination of objective factors X that affect the travel time length of the road section and labels assigned to the road section by the user, such as objective labels T.

[0026] As an example, the following linear equation can be employed to calculate the objective travel time Yi:

number

[0027] Here, β0 represents the ideal time length to pass through road section i, i.e., the time length to pass through the road section under ideal traffic conditions and traffic laws, and X jrepresents the quantitative value of the objective factors that affect the travel time of road section i, and these objective factors include, for example, road surface construction status, weather conditions, traffic accident rate, speed limit of the road section, speed measurement location of the road section, road level, etc., and these factors can be collected by the navigation system. j is a regression parameter, j represents the jth objective factor affecting road section i, and m represents the number of objective factors affecting road section i.

[0028] Also, T in the formula k represents the quantitative value of the objective factor T, where k = 1, 2, ... p, and indicates that there are p objective factors of type T in road section i. The product X of X and T is h T h represents the mutual influence of X and T, where h=1, 2, ... mp, and mp represents the number of T and X factors simultaneously present on road segment i.

[0029] In addition, the coefficient β of each factor j , α k , φ h is determined by combining past time length data AT, ET, etc. through regression analysis, and is determined based on maximum likelihood estimation using, for example, a linear multiple regression model that is often used in conventional technology, and ε is a random error term determined by regression analysis.

[0030] As can be seen from the above formula (1), the calculation of the objective travel time Yi includes the objective factors X that affect the transit time collected based on the navigation system, and determines the travel time of the road section i that is affected by these objective factors as follows:

number

[0031] Next, the label T is used to adjust the time, and in this example, the influence of the label T is expressed as follows:

number

[0032] As can be seen from the above equation, when considering the objective label T assigned to a road section, not only is the impact of the objective label T itself on travel time taken into account, but also the mutual influence of the label T and the objective factor X extracted by the system. It is easy to understand that the presence of both the user-assigned objective factor T and the actual objective factor X is highly likely to exacerbate the adverse impact on travel time. For example, for factor X, weather, if weather conditions worsen, the occurrence of cut-ins (i.e., objective factors of type T) on the road section will increase, making the road section even more difficult to pass through. Therefore, according to this embodiment of the present invention, the impact of the objective label T itself on travel time is taken into account by calculating a duration adjustment value related to the objective factor T, and the impact of the mutual influence of the label T and the objective factor X extracted by the system on travel time is taken into account by calculating a duration adjustment value related to the mutual influence of the objective factor T and the objective factor X. It should be noted here that the present invention does not necessarily require the addition of labels to all u road sections when calculating the objective travel time for each road section, and the objective travel time can be adjusted based on the road sections that have labels.

[0033] This embodiment may also include step 305, namely, further adjusting the objective travel time Yi, for example, the time determined based on equation (1) in the above example, based on the subjective sensation label S recorded for the road section i, thereby forming the subjective travel time (also referred to as "subjective travel time length" in this specification), hereinafter referred to as Zi, for the road section i. In this example, the objective travel time Yi is adjusted by calculating an adjustment variable λ related to the mutual influence of the objective factor T, the objective factor X, and the subjective factor S. For example, such mutual influence is represented by the product of the objective factor T, the objective factor X, and the subjective factor S, and the adjustment variable λ can be calculated, for example, by a logarithmic function, as follows:

number

[0034] where S l f(X l )·g(T l ) represents the mutual influence factors of S, X, and T, and mpq represents the number of simultaneous influences of the three factors S, X, and T, where S is the tolerance value of each subjective factor label in the road section. f() and g() are functions of X and T, respectively, for normalizing X and T according to the actual situation. The navigation system uses a semantic understanding model to determine whether there is mutual influence between S, X, and T according to the content of the user label. For example, if the user label is "I won't go if it rains" (factor S), the semantic model can understand that the label S has a mutual influence with whether it is raining (factor X). When the user plans a route, if the actual weather is indeed raining, f(X) can be set to a value greater than zero so that the S label is effective. For example, the heavier the forecast or actual rain is, the larger the value of f(x) should be set. If it is not raining, the S label can be ineffective and f(X) can be set to 0, i.e., the mutual influence or the influence of rain does not need to be considered. Alternatively, if the meaning of the user label is more precise, for example, "the heavier the rain, the less likely I am to go," and if it is actually raining, f(X) can be set according to the amount of rain, preferably greater than 1; the more rain, the larger the possible value of f(X). Similarly, the g() function can be understood using a semantic model to determine whether there is a mutual influence between label S and label T. If there is an influence, the g() function can be set to a value greater than zero, and the greater the influence, the larger the possible value, preferably greater than 1. If there is little or no influence, it can take a small value in the (0,1) interval, or even 0.

[0035] This allows the subjective travel time Zi for road section i to be calculated as follows:

number

[0036] It should be noted here that if there is no mutual influence between S and X or T, for example, if only subjective factor S exists in road section i and there are no objective factors X and T that may be affected, then the function f(x) = g(T) = 1 in equation (2), meaning that for road section i, the variables can be determined taking into account only the S factor, and the variable calculation formula for road section i is as follows:

number

[0037] If there is only X or T that interacts with S, for example, if there is only X, then g(T)=1, and equation (2) can be simplified as follows:

number

[0038] Next, in step 307, it is determined whether the subjective travel time Zi has been calculated for each of the u road sections in the current route R1. If not, the above steps are repeated until the subjective travel time Zi for all u road sections has been calculated, and then the process proceeds to step 311.

[0039] In step 309, the expected travel time length of the route R1 is calculated, and the expected travel time length is defined as the sum Z1 of the subjective travel times Zi of all the u road sections included therein. That is,

number

[0040] In this way, for step 301, three recommended routes R1, R2, and R3 can be determined, and the respective subjective travel times Z1, Z2, and Z3 can be calculated. After determining that the calculation of the expected travel time lengths of all the recommended routes has been completed in step 311, in step 313, the navigation system determines the route having the minimum value Z among the three recommended routes as the optimal route and recommends it to the user, that is, Z = min{Z1, Z2, Z3}. For example, when Z1 < Z2 < Z3, route R1 is recommended to the user as the optimal navigation travel route.

[0041] It should be pointed out here that in the above example, Yi is calculated by a linear function and the adjustment variable is calculated by a logarithmic function with base 2. However, as is clear, the present invention is not limited to this, and according to the actual situation, other calculation methods can be adopted or optimized, as long as the influence of each label factor on the travel time length can be reflected.

[0042] Also, in the above example, a specific exemplary embodiment is proposed by simultaneously considering the influence of the objective factor T and the subjective factor S added by the user on the travel time length. However, in other implementation modes of the present invention, only one type of factor added by the user is considered, for example, only the objective factor of type T or only the subjective factor of type S is considered to calculate the expected time length of the route. For example, when only factor T is considered, the expected time length Z of route R can be calculated based on the following formula.

Equation

[0043] Thereby, the navigation system recommends the route having the minimum value Z among each route R as the optimal route.

[0044] After determining the optimal navigation route in step 205 or 317, the user can start navigation. At the same time, the navigation system can ask the user whether they want to allow the system to record driving status information for each road segment, such as the duration of each road segment, and whether they are willing to make a note. Of course, the navigation system can automatically record driving information based on system settings and receive labels actively entered by the user or automatically generated by the system. As a result, during the vehicle's travel, the navigation system records the estimated duration and actual duration information for each road segment. If the actual duration is longer than the estimated duration, the navigation system can automatically generate a label or prompt the user to add one, for example, by popping up a dialog interface on the GUI screen and allowing the user to select an appropriate label from a drop-down menu. This allows the user's past data to be constantly updated in the database (DB), thereby providing better service to the user on subsequent trips.

[0045] Although the present invention has been described above as an exemplary embodiment, it is understood that the present invention is not limited thereto and can be modified. For example, if a tolerance value is assigned to the type S label, the road segments of the multiple navigation recommended routes determined in step 201 can be filtered based on the degree value, for example, road segments and their associated recommended routes having a degree value below a predetermined threshold can be removed, avoiding interference with routes with low tolerance, thereby further optimizing the candidate routes and road segments presented to the user by the GUI.

[0046] The above describes the implementation of route planning by the navigation system of the present invention using a specific example, where the navigation system is embedded in the in-vehicle infotainment system of a vehicle, and implements the route planning method of the present invention by executing a graphical user interface (GUI) and machine-readable program code by at least one processing unit. For example, FIG. 4 shows a schematic diagram of a navigation system according to an example of the present invention, where a user provides user input via a voice or text interface provided by the GUI, and at least one computing unit 200 executes computer-executable instructions to plan a driving route by implementing the method disclosed herein based on the user input and the historical data and label data stored in database 300. In one example, the computer-readable executable instructions can be stored in a local or remote storage medium. Therefore, the scope of protection of the present invention is limited by the claims. [Explanation of symbols]

[0047] 200 compute units 300 databases

Claims

1. Dividing each of the plurality of navigation recommended routes into at least one road segment; determining a navigation travel route from among the plurality of recommended navigation routes based on the user's past travel information in the at least one road section of each of the routes and label information added by the user; A navigation route planning method comprising:

2. the label information includes a first label customized by a user and indicating a first objective factor affecting transit time; The method comprises: determining a basic travel time for each road section in each navigation-recommended route based on a second objective factor that affects the travel time collected by the navigation system; and adjusting the base running time based on the first label to generate an objective running time.

3. adjusting the objective running time based on the first label, calculating a first duration adjustment value related to the first objective factor; calculating a second duration adjustment value related to the mutual influence of the first objective factor and the second objective factor; 3. The method of claim 2, further comprising adjusting the base running time using the first duration adjustment value and the second duration adjustment value to generate the objective running time.

4. the label information includes a second label customized by a user and indicating a subjective factor affecting the transit time; The method comprises: The method of claim 3 , further adjusting the objective travel time based on the second label to determine a subjective travel time for the road segment.

5. Further adjusting the objective running time based on the second label includes: calculating a moderator variable related to the mutual influence of the first objective factor, the second objective factor, and the subjective factor; 5. The method of claim 4, further comprising utilizing the adjustment variable to adjust the objective driving time to generate the subjective driving time.

6. 6. The method of claim 5, wherein the mutual influence of the first objective factor, the second objective factor, and the subjective factor is quantified by a product of a tolerance value representing the subjective factor, a first function representing the influence of the first objective factor, and a second function representing the influence of the second objective factor.

7. The method according to claim 5 , wherein the determined navigation travel route is a navigation-recommended route having a road section with the smallest sum of subjective travel times among the plurality of navigation-recommended routes.

8. 8. The method of claim 7, wherein the second label is classified as having multiple tolerance values, and different subjective factors belonging to the second label are assigned one of the multiple tolerance values.

9. The first objective factor includes one or more selected from the group consisting of a red light duration, a number of intersections, and a number of vehicles cutting in, The second objective factor includes one or more selected from the following: road surface construction status, weather conditions, traffic accident rate, speed limit of the road section, speed measurement location of the road section, and road level; The method of claim 1 , wherein the subjective factors include one or more selected from factors that challenge a driving level, including sudden events, narrow roads, and pedestrians.

10. the label information includes a first label customized by a user and indicating a first objective factor that affects the travel time of the road segment, and a second label customized by a user and indicating a subjective factor that affects the travel time of the road segment; The method comprises: The method of claim 1 , further comprising determining a navigation driving route from among the plurality of recommended navigation routes based on the past driving information and the first and second labels.

11. the second label is assigned one of a plurality of tolerance values; The method comprises: filtering the road segments from the plurality of navigation-recommended routes based on the tolerance values ​​assigned to the second labels to obtain filtered navigation-recommended routes; and selecting the navigation travel route from among the filtered recommended navigation routes.

12. displaying, via a user interface, past information for each road segment and the filtered road segment including the first label and / or the second label, wherein the past information includes past predicted travel time lengths and actual travel time lengths; 12. The method of claim 11, further comprising receiving user input on the user interface to select preferred road segments, wherein the navigation driving route includes the preferred road segments selected by the user.

13. 13. The method of claim 1, further comprising: in response to a user triggering a currently traveled or traversed road segment on the navigation travel route, receiving label information added by the user to the selected road segment, the label information reflecting objective or subjective factors that influence travel on the road segment, customized by the user.

14. 13. The method of claim 1, further comprising the step of providing a user interface for receiving label information added by a user to a selected road segment in response to an actual driving time length of the traversed road segment on the navigation driving route being greater than an expected driving time length, the label information reflecting objective or subjective factors that affect driving on the road segment, customized by the user.

15. a graphical user interface (GUI) configured to provide a user input interface for receiving user input in the form of voice, text, or touch; and at least one computing unit configured to perform the method according to any one of claims 1 to 14.

16. A computer program comprising computer readable program code which, when executed by a computing device, causes the computing device to carry out a method according to any one of claims 1 to 14.

Images