Guiding point acquisition method and device, equipment and storage medium

Abstract

The invention provides a guide point acquisition method and device, equipment and a storage medium, and relates to the field of artificial intelligence, in particular to the fields of map navigation, automatic driving, cloud computing, Internet of Things and intelligent transportation. According to the specific implementation scheme, the method comprises the steps of obtaining a candidate set associated with a point of interest based on associated data of the point of interest; wherein the candidate set comprises a plurality of candidate points; filtering the plurality of candidate points in the candidate set based on the orientation area of the interest point; and acquiring a guide point associated with the interest point based on the candidate points reserved after filtering.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to the fields of autonomous driving, cloud computing, the Internet of Things, and intelligent transportation. Background Art

[0002] Map navigation can provide path planning and navigation guidance from the starting location to the destination with the help of satellite positioning systems, map data and related algorithms. On the planned navigation route, you can set key location markers. These markers can guide users forward and can be called guide points. In order to guide users to reach their destination accurately, you can use guide points near the destination to guide users forward. Summary of the invention

[0003] The present disclosure provides a method, apparatus, device and storage medium for obtaining a guidance point.

[0004] According to one aspect of the present disclosure, there is provided a method for obtaining a guidance point, comprising:

[0005] Based on the associated data of the point of interest, a candidate set associated with the point of interest is obtained; wherein the candidate set includes a plurality of candidate points;

[0006] Based on the orientation area of ​​the point of interest, filtering the multiple candidate points in the candidate set;

[0007] Based on the candidate points retained after filtering, a guide point associated with the interest point is obtained.

[0008] According to another aspect of the present disclosure, there is provided a guidance point acquisition device, comprising:

[0009] A candidate set acquisition module, used to acquire a candidate set associated with the point of interest based on the associated data of the point of interest; wherein the candidate set includes multiple candidate points;

[0010] A filtering module, configured to filter the plurality of candidate points in the candidate set based on the orientation area of ​​the point of interest;

[0011] The guide point acquisition module is used to acquire the guide point associated with the interest point based on the candidate points retained after filtering.

[0012] According to another aspect of the present disclosure, there is provided an electronic device, comprising:

[0013] at least one processor; and

[0014] a memory communicatively connected to the at least one processor; wherein,

[0015] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute any method in the embodiments of the present disclosure.

[0016] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute any method according to the embodiments of the present disclosure.

[0017] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, which implements any method according to the embodiments of the present disclosure when executed by a processor.

[0018] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0020] Figure 1 is a flowchart of a method for obtaining a guidance point according to an embodiment of the present disclosure;

[0021] Figure 2 is a schematic diagram of a delivery point in an embodiment of the present disclosure;

[0022] Figure 3 is a schematic diagram of a binding point in an embodiment of the present disclosure;

[0023] Figure 4 is a flow chart of a method for obtaining a guidance point according to another embodiment of the present disclosure;

[0024] Figure 5 It is a schematic diagram of selecting candidate points according to the orientation area of ​​the interest point;

[0025] Figure 6 It is a schematic diagram of determining the direction of the triangular area according to the direction vector of the interest point;

[0026] Figure 7 It is a schematic diagram for determining the direction of the fan-shaped area based on the point of interest;

[0027] Figure 8 is a flow chart of a method for obtaining a guidance point according to another embodiment of the present disclosure;

[0028] Fig. 9 is a flow chart of a method for obtaining a guidance point according to another embodiment of the present disclosure;

[0029] Fig.10 is a flow chart of a method for obtaining a guidance point according to another embodiment of the present disclosure;

[0030] Fig.11 is a flow chart of a method for obtaining a guidance point according to another embodiment of the present disclosure;

[0031] Fig.12 is a flow chart of a guide point mining method based on navigation trajectory;

[0032] Fig.13 It is a flow chart of the mining process of public travel trajectory data;

[0033] Fig.14 It is a flowchart of candidate point filtering and sorting;

[0034] Fig.15 is a structural schematic diagram of a guidance point acquisition device according to an embodiment of the present disclosure;

[0035] Fig.16 is a structural schematic diagram of a guidance point acquisition device according to another embodiment of the present disclosure;

[0036] Fig.17 The block diagram is a block diagram of an electronic device for implementing the guidance point acquisition method according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0037] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0038] Figure 1 1 is a flow chart of a method 100 for obtaining a guidance point according to an embodiment of the present disclosure, the method comprising:

[0039] S110, based on the associated data of the point of interest, obtaining a candidate set associated with the point of interest; wherein the candidate set includes a plurality of candidate points;

[0040] S120, filtering the multiple candidate points in the candidate set based on the orientation area of ​​the point of interest;

[0041] S130: Based on the candidate points retained after filtering, obtain a guide point associated with the point of interest.

[0042] In the disclosed embodiments, various points of interest (POI) may be included on the navigation map. A POI may be used to represent a specific location on the map. For example, supermarkets, banks, hospitals, pharmacies, gas stations, restaurants, hotels, cinemas, parks, museums, historical sites, and other locations. In some scenarios, a POI may be the final destination that a navigation needs to go to, or it may be a nearby POI generated based on the input navigation destination. For example, if the input destination is a shopping mall, the POI may be a tea shop, coffee shop, and the like at the entrance of the shopping mall.

[0043] Each point of interest may have a unique point of interest identifier (ID) on the map, and each point of interest identifier may be bound to the associated data of the point of interest. If the point of interest is subsequently selected as the navigation destination, the guide point of the point of interest may be obtained based on the associated data of the point of interest. The guide point is usually located on a road network and can guide the user to the point of interest through the passable roads on the road network. The associated data of the point of interest may include at least one data such as a trajectory, an image, a road network, etc. near the point of interest. For example, the associated data of a certain point of interest includes at least one data such as a trajectory, an image, a road network, etc. within a specified range, such as 100 meters, from the point of interest. According to the associated data of the point of interest, all candidate points within a specified range, such as 100 meters, near the point of interest can be obtained, and all candidate points can be added to the candidate set of the point of interest (also referred to as a candidate point set). These candidate points can also be understood as optional guide points, that is, one or more guide points can be selected from these candidate points.

[0044] In the disclosed embodiments, a point of interest generally has a direction, such as the entrance direction of a building represented by the point of interest, the direction of a door opening, etc. The direction region of the point of interest can be obtained by extending outward from the entrance or opening based on the point of interest. The candidate point within the direction region of the point of interest is more suitable as a guide point for the point of interest. Therefore, the candidate points in the candidate set of a certain point of interest that are not within the direction region of the point of interest can be filtered out, and the candidate points in the candidate set that are within the direction region of the point of interest are retained.

[0045] In the disclosed embodiment, among the candidate points retained after filtering based on the orientation area of ​​the point of interest, one or more final guide points of the point of interest may be selected in one or more ways. For example, one or more guide points may be randomly selected. For another example, the guide point closest to the point of interest may be selected. For another example, a model may be used to score the retained candidate points, and one or more guide points with higher scores may be selected.

[0046] According to an embodiment of the present disclosure, a candidate set associated with the point of interest is obtained based on the associated data of the point of interest, and a guidance point is screened out from the candidate points of the candidate set based on the direction area of ​​the point of interest. The guidance point of the point of interest can be further determined among the candidate points in the direction area of ​​the point of interest, thereby obtaining a more reasonable and accurate guidance point, which can further improve the accuracy and precision of navigation.

[0047] In one implementation, S110 acquires a candidate set associated with the point of interest based on the associated data of the point of interest, including one or more of the following:

[0048] Obtaining a delivery point based on one or more of the public travel trajectory data, the driving trajectory data, and the road network data in the associated data, and adding the delivery point to the candidate set;

[0049] Obtaining a binding point based on one or more of the public travel trajectory data, the driving trajectory data, and the road network data in the associated data, and adding the binding point to a candidate set;

[0050] The trajectory points of the public travel trajectory data and / or the driving trajectory data in the associated data are added to the candidate set.

[0051] In the disclosed embodiments, the associated data of the points of interest may include public travel trajectory data, driving trajectory data, road network data, image data of the points of interest, etc. Among them, the public travel trajectory data may include public travel trajectory data generated by using public travel vehicles such as buses, subways, and shared bicycles, and may be based on public transportation application software, intelligent transportation systems, and other public travel trajectory data. The driving trajectory data may include driving trajectory data generated by driving a car, and may be based on driving trajectory data such as a driving recorder, a car computer, navigation software, and a vehicle tracking device.

[0052] In the disclosed embodiment, the route trajectory near the point of interest, for example, within a set range from the point of interest, can be obtained through public travel trajectory data or driving trajectory data. Figure 2 As shown, points A and B where the route track near the point of interest intersects with the link in the road network data can be added to the candidate set as intersection points as candidate points of the point of interest. Figure 3 As shown, a vertical line can be drawn from the point of interest toward the road network links near the point of interest, such as Link1 and Link2, and the vertical points C and D can be added to the candidate set as binding points as candidate points of the point of interest. For another example, part or all of the track points on the public travel track and driving track near the point of interest can also be added to the candidate set as candidate points of the point of interest.

[0053] In the disclosed embodiments, a road network link (Link) may include a road segment connecting two nodes of a road network (such as an intersection, a station, an end point, or other important geographical locations). These road segments are the basic units that constitute a road network and can be used to represent the connectivity of roads. The topological relationship of a road network can be described by a road network link (Link). For example, the section of a road from one intersection to the next intersection can be regarded as a road network link "Link".

[0054] In the above example, the road network link may include a road-level link in the road network, or may include a lane-level link in the road network. The lane-level link may be used to more accurately determine the lane-level guidance point.

[0055] The track points in the above example may include road-level track points or lane-level track points. Lane-level track points may be used to more accurately determine lane-level guide points.

[0056] According to the embodiments of the present disclosure, by selecting candidate points of interest in a variety of ways, the accuracy of the finally generated guidance points can be improved, thereby improving the navigation effect.

[0057] Figure 4 4 is a flow chart of a method 400 for obtaining a guide point according to another embodiment of the present disclosure, and the method 400 can be used to implement step S120 in the method 100 for obtaining a guide point. In one embodiment, the method 400 includes: filtering the multiple candidate points in the candidate set based on the direction area of ​​the point of interest, and further includes:

[0058] S410, calculating the direction area of ​​the point of interest based on the associated data;

[0059] S420: Remove candidate points that are not in the direction area of ​​the point of interest from the candidate set to obtain a retained first candidate point.

[0060] In the disclosed embodiment, the map may include basic orientation information of each point of interest. For example, one opening of the point of interest faces east, and one opening faces west, etc. The orientation area of ​​the point of interest may be determined based on the basic orientation information of the point of interest and the associated data of the point of interest, such as public travel trajectory data, driving trajectory data, image data, road network link data, and sign library data near the point of interest. The orientation area may include an area extending outward from the opening of the point of interest for a certain length.

[0061] In the embodiment of the present disclosure, based on the direction area of ​​the point of interest, candidate points that are not in the direction area of ​​the point of interest can be removed from the candidate set of the point of interest. Figure 5As shown, the orientation area of ​​the interest point includes candidate points A, B and C, but does not include candidate point D. Then candidate point D is removed, and candidate points A, B and C are retained. An interest point may include one or more orientation areas. For example, the interest point has two openings. Opening 1 corresponds to one orientation area, and opening 2 corresponds to another orientation area.

[0062] According to the embodiments of the present disclosure, by filtering candidate points according to the direction area of ​​the point of interest, the amount of calculation can be reduced, the accuracy of the guidance points can be improved, and the accuracy and precision of navigation can be improved.

[0063] In one embodiment, calculating the orientation area of ​​the point of interest based on the associated data includes: determining the orientation vector of the point of interest or the orientation fan-shaped area of ​​the point of interest based on one or more of the public travel trajectory data, driving trajectory data, point of interest image data, road network link data of the point of interest orientation, and sign library data in the associated data; wherein the orientation vector is used to determine the orientation triangle area of ​​the point of interest.

[0064] In the disclosed embodiment, the point of interest orientation prediction model can be used to predict the precise orientation vector of the point of interest based on one or more of the public travel trajectory data, driving trajectory data, image data of the point of interest, road network link data of the point of interest orientation, and sign library data associated with the point of interest. The road network link of the point of interest orientation is determined based on the orientation vector. Then, based on the point of interest and the road network link of the point of interest orientation, the orientation triangle area of ​​the point of interest is determined. Figure 6 As shown, from the opening position S1 of the point of interest, the road network link Link1 that the point of interest is facing can be determined along the orientation vector, and the orientation triangle area of ​​the point of interest is constructed using S1 and the two endpoints L1 and L2 of Link1, and the vertices of the triangle area are S1, L1 and L2.

[0065] In the embodiment of the present disclosure, if an accurate orientation vector cannot be obtained, the orientation prediction model of the point of interest can predict a fuzzy orientation sector. The orientation sector can include a sector from the opening position S1 of the point of interest to the road link closer to the position of the point of interest, such as Figure 7 As shown, the endpoints of the sectors may include S1, L3, and L4.

[0066] In the disclosed embodiment, the POI direction prediction model can be obtained by fine-tuning based on the basic large model using training samples with data types similar to those included in the associated data.

[0067] According to the embodiments of the present disclosure, various associated data can be used to predict the direction area of ​​the point of interest, which helps to accurately screen the candidate points of the point of interest, reduce the amount of calculation, improve the accuracy of the guidance points, and thus improve the accuracy and precision of navigation.

[0068] Figure 8 8 is a flow chart of a method 800 for obtaining a guide point according to another embodiment of the present disclosure, and the method 800 can be used to implement step S130 in the method 100 for obtaining a guide point. In one embodiment, the method 800 includes: obtaining a guide point associated with the point of interest based on the candidate points retained after filtering, and further includes:

[0069] S810, sorting the plurality of first candidate points based on the features of the plurality of retained first candidate points to obtain a first sorting result;

[0070] S820: Select a first number of second candidate points from the first sorting result.

[0071] In the disclosed embodiment, the first candidate point is a candidate point retained after filtering in the candidate set, and the first candidate point retained in the candidate set may be one or more. Sorting parameters may be calculated based on the features of the first candidate point, and the sorting parameters may include a sorting score, a sorting probability value, and the like. Generally speaking, the higher the score or probability value of the candidate point, the closer the relationship between the candidate point and the point of interest, and the higher the quality, and the reverse setting may also be possible. The features of the candidate point may include various distances between the candidate point and the point of interest, information about the road network where the candidate point is located, trajectory information related to the candidate point, and the like.

[0072] According to the embodiment of the present disclosure, the retained candidate points are sorted based on their features, and a specific number of candidate points are selected from the sorting results, so that more accurate candidate points can be further screened out, the accuracy of the guidance points can be improved, and the accuracy and precision of navigation can be improved.

[0073] In one embodiment, the characteristics of the first candidate point include one or more of the following: the straight-line distance to the point of interest, the walking distance to the point of interest, the ratio of the walking distance to the straight-line distance, the traffic volume of the road network link, the road network grade of the road network link, the traffic volume of public travel trajectory data, and the traffic volume of driving trajectory data.

[0074] In the disclosed embodiments, in the features of the candidate points, the distance between the candidate point and the point of interest may include the explicit straight-line distance between the first candidate point and the point of interest. For example, the straight-line distance AS from the first candidate point A to the point of interest S, the explicit straight-line distance may also be referred to as the aerial straight-line distance. In the features of the candidate points, the distance between the candidate point and the point of interest may include the actual walking distance between the first candidate point and the point of interest. For example, the length of each path taken by walking from the first candidate point A to the point of interest S, the actual walking distance may also be referred to as the ground walking distance. The actual walking distance may be obtained using the walking navigation mode of the navigation software.

[0075] In the disclosed embodiment, in the characteristics of the candidate point, the information of the road network where the candidate point is located may include the traffic volume of the road network link where the first candidate point is located, for example, it may include the number of various tracks passing through the road network link where the first candidate point is located. In the characteristics of the candidate point, the information of the road network where the candidate point is located may include the road network level of the road network link where the first candidate point is located. For example, the road network level of a road network link may be an expressway, a main road, a secondary road, or a branch road. For another example, the road network level of a road network link may be primary, secondary, tertiary, or quaternary, etc.

[0076] In the disclosed embodiment, in the features of the candidate points, the trajectory information related to the candidate points may include the flow of public travel trajectory data and / or driving trajectory data passing through the first candidate point. For example, the number of public travel trajectories passing through candidate point B is 20, and the number of driving trajectories is 10.

[0077] By inputting one or more of the features of the above candidate points into the sorting model, the sorting result of the first candidate point can be calculated. The sorting model can be a basic model such as a Light Gradient Boosting Machine (Light GBM) for sorting. Light GBM is a tree-based ensemble learning model that uses gradient boosting technology. By combining multiple weak learners (usually decision trees) into a powerful model, data samples can be sorted, classified or regressed. The sorting model can also be a model obtained by fine-tuning the basic sorting model based on training samples related to the features.

[0078] In addition, the set sorting algorithm can also be used to calculate the sorting result. For example, a simple sorting algorithm can be a weighted sorting algorithm, which converts various features into scores first, then multiplies them with their corresponding weights, and finally adds them up to get the total score of the candidate point. For example, the score converted from the straight-line distance to the point of interest is S1, and its corresponding weight is W1; the score converted from the walking distance to the point of interest is S2, and its corresponding weight is W2; the score converted from the ratio of walking distance to straight-line distance is S3, and its corresponding weight is W3; the score converted from the flow of the road network link where the first candidate point is located is S4, and its corresponding weight is W4; the score converted from the flow of the public travel trajectory data passing through the first candidate point is S5, and its corresponding weight is W5. In this case, the total score of the candidate point = S1×W1+S2×W2+S3×W3+S4×W4+S5×W5. Examples of formulas for selecting other features can be modified by referring to the above examples.

[0079] There are many ways to convert various features into scores. One example may include: setting scores corresponding to different parameter ranges of the feature, and finding the corresponding scores according to the specific numerical values ​​of the feature. For example, the straight-line distance from the first candidate point to the point of interest is less than 10 meters and is scored as 10, between 10 meters and 20 meters and is scored as 9, and between 20 meters and 30 meters and is scored as 8. If the straight-line distance AS from the first candidate point A to the point of interest S is 25 meters, the score is 8. Other examples of converting features into scores can refer to the above examples for modification.

[0080] In the embodiment of the present disclosure, an example of the first sorting result may include: sorting from front to back in order of the scores of the first candidate points from high to low. For example, the first sorting result includes candidate points {A, C, B, F, E...}. According to the value of the first number to be selected, the first number of candidate points can be selected in the order from front to back in the first sorting result. For example, if the first number is 4, the second candidate points selected from the above first sorting result include {A, C, B, F}.

[0081] According to the embodiment of the present disclosure, by calculating the features of the first candidate point based on the multiple features of the first candidate point, the accuracy of the sorting result can be improved, thereby improving the accuracy of the guide point.

[0082] In one embodiment, the road network link includes a lane-level link in the road network. For example, the features of the first candidate point used for sorting may include the traffic of the lane-level link where the first candidate point is located. For example, if the first candidate point is on the lane-level link Link1-1 of the road network link Link1, the number of various trajectories passing on the lane-level link Link1-1 can be used as a parameter to calculate the sorting result. The use of lane-level links can more accurately determine the lane-level guidance point, so that the vehicle can travel to the guidance point more quickly and accurately along the recommended lane.

[0083] In one embodiment, the driving trajectory data includes lane-level driving trajectory data. For example, the features of the first candidate point used for sorting may include lane-level driving trajectory data passing through the first candidate point. For example, the first candidate point is on the lane-level link Link1-1 of the road network link Link1, and the number of driving trajectories passing through the lane-level link Link1-1 is 8. The use of lane-level links can more accurately determine the lane-level guidance point, so that the vehicle can drive to the guidance point more quickly and accurately along the recommended lane.

[0084] Fig. 9is a flow chart of a method 900 for obtaining a guide point according to another embodiment of the present disclosure, and the method 900 can be used to implement step S130 in the method 100 for obtaining a guide point. In one embodiment, the method 900 includes: obtaining a guide point associated with the point of interest based on the candidate points retained after filtering, and further includes:

[0085] S910, clustering the first number of second candidate points to obtain a clustering result, where the clustering result includes a third candidate point corresponding to each cluster center;

[0086] S920: Determine a guide point for the point of interest based on the third candidate point.

[0087] In an embodiment of the present disclosure, the second candidate point can be processed by a clustering algorithm. Clustering can divide a data set into different clusters according to a specific standard (such as distance), so that the similarity of data objects in the same cluster is as large as possible, and the difference of data objects not in the same cluster is as large as possible. The first number of second candidate points can be clustered using a clustering algorithm, and the obtained clustering results can include one or more clustering clusters. From each clustering cluster of the clustering results, the second candidate point can be screened again to obtain a third candidate point. If only one third candidate point is obtained, the third candidate point can be directly determined as the guide point of the point of interest. If multiple third candidate points are obtained, one or more guide points of the point of interest can be selected again from the multiple third candidate points.

[0088] According to the embodiment of the present disclosure, the accuracy of the candidate points can be further improved by screening the candidate points again through the clustering algorithm, thereby improving the accuracy of the guide points.

[0089] Fig.10 1 is a flow chart of a method 1000 for obtaining a guide point according to another embodiment of the present disclosure, and the method 1000 can be used to implement step S910 in the method 900 for obtaining a guide point. In one embodiment, the method 900 includes: performing clustering processing on the first number of second candidate points to obtain a clustering result, and further includes:

[0090] S1010, clustering the first number of second candidate points based on a K nearest neighbor algorithm to obtain one or more clusters;

[0091] S1020: Use the second candidate point closest to the cluster center of the one or more clusters as the third candidate point.

[0092] In the disclosed embodiment, the K-Nearest Neighbor (KNN) algorithm is a basic classification and regression algorithm. In the KNN algorithm, for a new data point, the K data points (i.e., nearest neighbors) that are most similar to it can be found in the data set, and then the category or value of the new data point is predicted based on the category (in the classification problem) or value (in the regression problem) of these K neighbors. Based on the KNN algorithm, the second candidate point can be clustered to obtain one or more clusters and the cluster center of each cluster. KNN is only an example of a clustering algorithm, and other clustering algorithms can also be used, which can be flexibly selected according to the needs of the actual application scenario.

[0093] In the disclosed embodiment, in each cluster, the second candidate point closest to the cluster center can be selected as the third candidate point of the cluster. If there is only one cluster, a third candidate point can be obtained based on the cluster center, and the third candidate point can be directly determined as the guide point of the point of interest. If there are multiple clusters, multiple third candidate points can be obtained based on the cluster center, and one or more guide points of the point of interest can be selected again from the multiple third candidate points.

[0094] According to the embodiment of the present disclosure, the accuracy of the candidate points can be further improved by screening the candidate points again through the clustering algorithm, thereby improving the accuracy of obtaining the guide points.

[0095] Fig.11 1 is a flow chart of a method 1100 for obtaining a guide point according to another embodiment of the present disclosure, and the method 1100 can be used to implement step S920 in the method 900 for obtaining a guide point. In one embodiment, the method 900 includes: determining a guide point for the point of interest based on the third candidate point, and further includes:

[0096] S1110, sorting multiple third candidate points to obtain a second sorting result;

[0097] S1120: Select, from the second sorting result, a third candidate point that is closest to the point of interest as a guide point.

[0098] In the embodiment of the present disclosure, the plurality of third candidate points selected from the plurality of cluster centers may be sorted according to their respective distances from the point of interest. The second sorting result may be obtained by sorting by a sorting model or directly sorting by distance.

[0099] In one method, the second sorting result can be retained as a whole, and in application scenarios such as navigation, one or more candidate points can be provided according to the second sorting result point of interest based on user needs. For example, assume that the third candidate points selected from multiple second candidate points {A, C, B, F} include {A, C}. Among them, the distance between C and the point of interest is the smallest, and the second sorting result can be {C, A}. In application scenarios such as navigation, if the second sorting result corresponding to the destination (i.e., the point of interest) is {C, A}, the user can be prompted that the destination can be reached from both position C and position A, and priority is given to recommending to go from position C to the destination.

[0100] In another way, the third candidate point that is the most forward from the second sorting result can also be selected as the guide point of the point of interest, and the other candidate points can be ignored or reserved. For example, in application scenarios such as navigation, if the second sorting result corresponding to the destination (i.e., the point of interest) is {C, A}, and position C is selected as the guide point, the user can be prompted to reach the destination from position C. If the user asks whether there are other routes, the user is prompted to reach the destination from position A.

[0101] According to the embodiment of the present disclosure, multiple third candidate points are sorted and screened to obtain a final guidance point, which can improve the accuracy of the guidance point and further improve the accuracy and precision of navigation.

[0102] In one implementation, the guidance point acquisition method of any embodiment of the present disclosure further includes:

[0103] Acquire data to be processed from multiple data sources; wherein the data to be processed includes one or more of the following: public travel trajectory data, driving trajectory data, road network data, point of interest orientation data, point of interest image data, and sign library data;

[0104] A corresponding relationship is established between the associated data in the to-be-processed data whose positional relationship with each point of interest meets the set requirements and the identifier of each point of interest.

[0105] In the disclosed embodiments, the data to be processed may be obtained from multiple data sources. These data sources may include, but are not limited to, navigation software, maps, road networks, Internet of Vehicles, sign libraries, etc. These data to be processed include, but are not limited to, public travel trajectory data, driving trajectory data, point of interest orientation data, point of interest image data, road network data, sign library data, etc. Among them, the public travel trajectory may include, for example, the trajectory data of public travel tools such as buses, walking, and cycling. The point of interest image data may include the image data of vehicles shot near the point of interest, the image data of smart cameras arranged near the point of interest, etc.

[0106] The method of preprocessing these data to be processed may include: associating the data to be processed near each point of interest with the identifier of the point of interest. The identifier of the point of interest may include the name and number of the point of interest, etc., which are used to uniquely identify a point of interest. According to the positional relationship between various data to be processed and a certain point of interest, one or more of public travel trajectory data, driving trajectory data, point of interest orientation data, point of interest image data, road network data, and sign library data whose positional relationship with the point of interest meets the set requirements (for example, the distance to the point of interest is within 200 meters) can be screened from the data to be processed as the associated data of the point of interest.

[0107] According to the embodiment of the present disclosure, data from multiple data sources are associated with a point of interest, so that a guide point for the point of interest can be subsequently determined based on data that is highly associated with the point of interest, thereby improving the accuracy of the guide point and further improving the accuracy of navigation.

[0108] In one implementation, the guidance point acquisition method of any embodiment of the present disclosure further includes: when the associated data includes public travel trajectory data, performing a denoising operation and / or a thinning operation on the public travel trajectory data.

[0109] In an embodiment of the present disclosure, before executing S110, it may be determined whether the associated data of a certain point of interest includes public travel trajectory data. If so, a denoising operation and / or a thinning operation may be performed on the public travel trajectory data in the associated data of the point of interest. Among them, the denoising operation may remove noise data and retain the authenticity and integrity of the public travel trajectory data as much as possible. The suction operation may sample public travel trajectories with a high density and simplify the public travel trajectory data as much as possible to reduce the amount of data that needs to be processed subsequently. The public travel trajectory data used in and after S110 may be a denoised and / or thinned public travel trajectory.

[0110] According to the embodiments of the present disclosure, the amount of public travel trajectory data can be reduced through denoising operations and / or suction operations, and reasonable public travel trajectories can be retained, thereby reducing the overall amount of calculation and improving the efficiency of obtaining guidance points.

[0111] In some examples, the destination guidance of map products can simply infer the location of the destination guidance point based on driving trajectory, images, road network and other data. The road network method can use the point of interest data and the road network to generate a series of candidate points, and then determine the driving trajectory characteristics on the candidate points, and obtain the final guidance point through voting. The image method can use the location of the destination to select multiple nearby images, determine the coordinates and shooting angle of the image to obtain the direction of the destination, and then find other images on the direction vector, and fuse these images to obtain the guidance point of the point of interest.

[0112] In daily life, public transportation accounts for a large part of all transportation modes. For example, buses, subways, walking, cycling, etc. are part of public transportation. Similarly, public transportation can reduce emissions and is conducive to sustainable development. However, in public transportation, there is a problem of destination arrival. Due to the variety of destination types and forms, including indoor and outdoor, large and small areas, it is very difficult to clearly determine the actual destination arrival point.

[0113] Due to the differences between public travel and driving navigation, the destination guidance points with driving characteristics obtained based on the road network and driving trajectory method may not be applicable to all travel modes. The disclosed embodiment can better utilize the public travel trajectory characteristics to characterize the user's real destination judgment point. In the image-based method, vehicle network images can be collected. If the coverage rate of other sources is very low, it cannot meet the practical requirements. In addition, the location of the car is generally on the main road. If the final guidance point is very different from the actual destination (point of interest) door, it will cause great trouble to the user.

[0114] The method for obtaining a guide point provided by the present disclosure includes a method for mining a guide point based on a navigation trajectory. The method can combine the navigation trajectory of public transportation such as buses, subways, walking, and cycling, and comprehensively utilize the public transportation trajectory, driving trajectory (driving production point), image information such as destination signs and storefronts, relationship information of map points of interest, and other data that can assist in determining the location of the destination point, and use a series of strategies to comprehensively determine and generate the destination guide point that the user really needs. The guide point can be a destination guide point dedicated to public transportation.

[0115] A flowchart of a guide point mining method based on navigation trajectory is shown in Fig.12 As shown:

[0116] 1. Data preprocessing: In this step, it can specifically include: S1201, integration of data from sources of points of interest: It is necessary to integrate all required data from different sources, including public travel trajectory (bus, walking, cycling) data (referred to as bus walking and cycling trajectory data), driving trajectory data, point of interest orientation data, image data such as car shots near points of interest, road network data, sign library data, etc. Then execute S1202 to determine the associated data corresponding to the point of interest identifier (ID), such as a key-value (KV) table. Specifically, the unique ID of the point of interest is linked to the corresponding source data through association to obtain a dependent data file for subsequent recall sorting logic processing. In addition, it can be routinely updated according to time. Generally, the trajectory takes a week of data aggregation, and some data can also be updated on a daily basis, that is, once a day.

[0117] 2. Candidate point recall: After obtaining the associated data, each point of interest ID can be processed separately. For each point of interest, check whether there is necessary public travel trajectory data. If not, it degenerates to be consistent with the driving guide point. If it exists, the trajectory is first denoised and thinned. Then in S1203, the road network intersection point is calculated and the candidate point is recalled. Specifically, the public travel trajectory data such as bus trajectory, walking trajectory, cycling trajectory, etc. near the point of interest can be taken, and the intersection point can be calculated with the nearby road network data to add to the candidate set; the binding points of the point of interest to the nearby road network can also be recalled and added to the candidate set. Optionally, some or all of the trajectory points in the public travel trajectory data and / or driving trajectory data near the point of interest can also be added to the candidate points.

[0118] For example, a process of mining public travel trajectory data, such as Fig.13 As shown. In S1301, session matching, recall and other operations can be performed based on bus, walking and cycling service logs, and navigation trajectories of public transportation. In S1302, denoising, thinning and other operations can be performed on the trajectory points in the trajectory data. In S1303, the trajectory near the starting point and / or end point (i.e., point of interest) can be truncated, aggregated and other processes. In S1304, the processed trajectory is matched with the point of interest ID. Then, in S1305, the intersection point calculation, binding point recall and other operations are performed in combination with the road network data. Finally, in S1306, mining continues until the guide point is produced.

[0119] 3. Feature calculation and filtering and sorting: In S1204, orientation filtering and candidate point feature calculation can be performed: Specifically, Fig.14As shown, after entering the candidate point filtering and sorting in S1401, the trajectory and the candidate point set recalled by the bound road can be combined in S1402 to calculate the direction of the point of interest. For example, the navigation trajectory such as the public travel trajectory, the driving trajectory, the image such as the license plate image, the road network ID such as the road network link (Link) of the point of interest, the sign library data and other comprehensive calculations are used to obtain the precise direction vector or fuzzy direction fan-shaped area of ​​the point of interest. Among them, the Link of the point of interest can be found according to the precise direction vector, and the point of interest and the Link can be used to obtain the direction triangle area. In S1403, it can be determined whether each candidate point of the candidate set is in the above-mentioned fan-shaped area or triangle area. In S1404, the candidate points that are not in the fan-shaped area or triangle area are eliminated. In S1405, the retained candidate points are added to the retained set. Then all candidate points that are not within the corresponding direction area are filtered. In S1406, features of data from other sources can be collected to calculate the features of each candidate point, such as the explicit straight-line distance to the point of interest, the actual walking distance to the point of interest, the ratio of walking distance to straight-line distance, the traffic volume (Passenger Volume, PV) of the link, the road network level of the link, whether it crosses the road, whether it is prohibited, the PV of public travel and the PV of driving trajectory, etc. In S1407 (also S1205), the candidate points can be sorted by tree model based on these features. For example, by inputting these features into a sorting model such as LightGBM, candidate points with high rankings can be obtained. For example, in S1408, the sorting model can score the candidate points according to their features, sort them according to the scores, and then filter and output the data of the top-ranked candidate points, such as the Top 10.

[0120] 4. Deduplication and guide point generation: In S1206, the sorted candidate points, such as all the 10 candidate points mentioned above, are subjected to KNN clustering deduplication. Then in S1207, the points closest to the cluster centers are retained, and finally the point with the highest relative order, such as TOP1, is output as the guide point generated for the point of interest. Multiple candidate points can also be retained as multiple guide points for the point of interest.

[0121] The solution of the disclosed embodiment can effectively mine the trajectory characteristics of public travel, thereby using the trajectory characteristics of public travel to predict the candidate locations of the guide points to a specific map point of interest, and then by combining the information of the point of interest itself, the road network information, etc. to assist in judging the quality of each candidate point, so as to sort and obtain accurate user arrival guide points. This has a better navigation effect for those remote, hard-to-find street shops and points of interest in large areas, and is expected to completely solve the arrival problem in scenarios such as the last 100 meters.

[0122] Fig.15is a structural diagram of a guidance point acquisition device 1500 according to an embodiment of the present disclosure. The device 1500 may include:

[0123] The candidate set acquisition module 1510 is used to acquire a candidate set associated with the interest point based on the associated data of the interest point; wherein the candidate set includes multiple candidate points;

[0124] A filtering module 1520, configured to filter the plurality of candidate points in the candidate set based on the orientation area of ​​the interest point;

[0125] The guide point acquisition module 1530 is used to acquire the guide point associated with the interest point based on the candidate points retained after filtering.

[0126] Fig.16 1 is a schematic diagram of the structure of a guidance point acquisition device 1600 according to another embodiment of the present disclosure, the device 1600 includes: a candidate set acquisition module 1610, a filtering module 1620 and a guidance point acquisition module 1630. The functions of the above modules can refer to the functions of the modules of the guidance point acquisition device in the above embodiment. In one embodiment, the candidate set acquisition module 1610 includes one or more of the following submodules:

[0127] The delivery point submodule 1611 is used to obtain a delivery point based on one or more of the public travel trajectory data, the driving trajectory data and the road network data in the associated data, and add the delivery point to the candidate set;

[0128] A binding point submodule 1612 is used to obtain a binding point based on one or more of the public travel trajectory data, the driving trajectory data and the road network data in the associated data, and add the binding point to a candidate set;

[0129] The trajectory point submodule 1613 is used to add the trajectory points of the public travel trajectory data and / or the driving trajectory data in the associated data to a candidate set.

[0130] In one embodiment, the filtering module 1620 includes:

[0131] A calculation submodule 1621 is used to calculate the direction area of ​​the point of interest based on the association data;

[0132] The processing submodule 1622 is used to remove the candidate points that are not in the direction area of ​​the point of interest from the candidate set, so as to obtain the retained first candidate point.

[0133] In one embodiment, the calculation submodule 1621 is used to predict the orientation vector of the point of interest or the orientation fan area of ​​the point of interest based on one or more of the public travel trajectory data, driving trajectory data, point of interest image data, road network link data of the point of interest orientation, and sign library data in the associated data; wherein the orientation vector is used to determine the orientation triangle area of ​​the point of interest.

[0134] In one implementation, the guidance point acquisition module 1630 includes:

[0135] A sorting submodule 1631 is used to sort the multiple first candidate points based on the features of the multiple first candidate points retained to obtain a first sorting result;

[0136] The selection submodule 1632 is configured to select a first number of second candidate points from the first sorting result.

[0137] In one embodiment, the characteristics of the first candidate point include one or more of the following: the straight-line distance to the point of interest, the walking distance to the point of interest, the ratio of the walking distance to the straight-line distance, the traffic volume of the road network link, the road network grade of the road network link, the traffic volume of public travel trajectory data, and the traffic volume of driving trajectory data.

[0138] In one embodiment, the road network link includes a lane-level link in the road network.

[0139] In one implementation, the driving trajectory data includes lane-level driving trajectory data.

[0140] In one implementation, the guidance point acquisition module 1630 includes:

[0141] A clustering submodule 1633 is used to perform clustering processing on the first number of second candidate points to obtain a clustering result, wherein the clustering result includes a third candidate point corresponding to each cluster center;

[0142] The determination submodule 1634 is configured to determine a guide point of the point of interest based on the third candidate point.

[0143] In one embodiment, the clustering submodule 1633 is used to cluster the first number of second candidate points based on the K nearest neighbor algorithm to obtain one or more clusters; and use the second candidate point closest to the cluster center of the one or more clusters as the third candidate point.

[0144] In one implementation, the determination submodule 1634 is used to sort the multiple third candidate points to obtain a second sorting result; and select the third candidate point closest to the point of interest from the second sorting result as the guide point.

[0145] In one embodiment, the apparatus 1600 further includes:

[0146] The data acquisition module 1640 is used to acquire data to be processed from multiple data sources; wherein the data to be processed includes one or more of the following: public travel trajectory data, driving trajectory data, road network data, point of interest orientation data, point of interest image data and sign library data;

[0147] The association module 1650 is used to establish a corresponding relationship between the associated data whose positional relationship with each interest point in the data to be processed meets the set requirements and the identifier of each interest point.

[0148] In one embodiment, the apparatus 1600 further includes:

[0149] The data processing module 1660 is used to perform a denoising operation and / or a thinning operation on the public travel trajectory data when the associated data includes the public travel trajectory data.

[0150] For the description of specific functions and examples of each module and submodule of the device in the embodiment of the present disclosure, reference can be made to the relevant description of the corresponding steps in the above method embodiment, which will not be repeated here.

[0151] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0152] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0153] Fig.17 A schematic block diagram of an example electronic device 1700 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0154] like Fig.17As shown, the device 1700 includes a computing unit 1701, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1702 or a computer program loaded from a storage unit 1708 into a random access memory (RAM) 1703. In the RAM 1703, various programs and data required for the operation of the device 1700 can also be stored. The computing unit 1701, the ROM 1702, and the RAM 1703 are connected to each other via a bus 1704. An input / output (I / O) interface 1705 is also connected to the bus 1704.

[0155] A number of components in the device 1700 are connected to the I / O interface 1705, including: an input unit 1706, such as a keyboard, a mouse, etc.; an output unit 1707, such as various types of displays, speakers, etc.; a storage unit 1708, such as a disk, an optical disk, etc.; and a communication unit 1709, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1709 allows the device 1700 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0156] The computing unit 1701 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1701 performs the various methods and processes described above, such as the guidance point acquisition method. For example, in some embodiments, the guidance point acquisition method may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 1708. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 1700 via the ROM 1702 and / or the communication unit 1709. When the computer program is loaded into the RAM 1703 and executed by the computing unit 1701, one or more steps of the guidance point acquisition method described above may be performed. Alternatively, in other embodiments, the computing unit 1701 may be configured to execute the guidance point acquisition method in any other appropriate manner (for example, by means of firmware).

[0157] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0158] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0159] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0160] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0161] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0162] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0163] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0164] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for obtaining a guide point, comprising: Based on the associated data of the point of interest, a candidate set associated with the point of interest is obtained; wherein the candidate set includes a plurality of candidate points; Based on the orientation area of ​​the point of interest, filtering the multiple candidate points in the candidate set; Based on the candidate points retained after filtering, a guide point associated with the point of interest is obtained.

2. The method according to claim 1, wherein: Based on the associated data of the point of interest, obtaining a candidate set associated with the point of interest includes one or more of the following: Obtaining a delivery point based on one or more of the public travel trajectory data, the driving trajectory data, and the road network data in the associated data, and adding the delivery point to a candidate set; Obtaining a road-binding point based on one or more of the public travel trajectory data, the driving trajectory data, and the road network data in the associated data, and adding the road-binding point to a candidate set; The trajectory points of the public travel trajectory data and / or the driving trajectory data in the associated data are added to a candidate set.

3. The method according to claim 1 or 2, wherein: Based on the orientation area of ​​the point of interest, filtering the multiple candidate points in the candidate set includes: Calculating the direction area of ​​the point of interest based on the associated data; The candidate points that are not in the direction area of ​​the point of interest are removed from the candidate set to obtain a retained first candidate point.

4. The method according to claim 3, wherein: Calculating the direction area of ​​the point of interest based on the association data includes: According to one or more of the public travel trajectory data, driving trajectory data, point of interest image data, road network link data of the point of interest orientation, and sign library data in the associated data, the orientation vector of the point of interest or the orientation fan area of ​​the point of interest is predicted; wherein the orientation vector is used to determine the orientation triangle area of ​​the point of interest.

5. The method according to any one of claims 1 to 4, wherein: Based on the candidate points retained after filtering, obtaining a guide point associated with the point of interest, including: sorting the plurality of first candidate points based on the features of the plurality of retained first candidate points to obtain a first sorting result; A first number of second candidate points are selected from the first sorting results.

6. According to the method according to claim 5, the characteristics of the first candidate point include one or more of the following: the straight-line distance to the point of interest, the walking distance to the point of interest, the ratio of the walking distance to the straight-line distance, the traffic volume of the road network link, the road network grade of the road network link, the traffic volume of public travel trajectory data and the traffic volume of driving trajectory data.

7. The method according to claim 6, wherein: The road network link includes a lane-level link in the road network, and the driving trajectory data includes lane-level driving trajectory data.

8. The method according to any one of claims 5 to 7, wherein: Based on the candidate points retained after filtering, obtaining a guide point associated with the point of interest, including: Performing clustering processing on the first number of second candidate points to obtain a clustering result, wherein the clustering result includes a third candidate point corresponding to each cluster center; Based on the third candidate point, a guide point for the point of interest is determined.

9. The method according to claim 8, wherein: Clustering the first number of second candidate points to obtain a clustering result includes: Clustering the first number of second candidate points based on a K nearest neighbor algorithm to obtain one or more clusters; The second candidate point closest to the cluster center of the one or more clusters is used as the third candidate point.

10. The method according to claim 8 or 9, wherein: Determining a guide point for the point of interest based on the third candidate point includes: Sorting the plurality of third candidate points to obtain a second sorting result; From the second sorting results, a third candidate point that is closest to the point of interest is selected as a guide point.

11. The method according to any one of claims 1 to 10, further comprising: Acquire data to be processed from multiple data sources; wherein the data to be processed includes one or more of the following: public travel trajectory data, driving trajectory data, road network data, point of interest orientation data, point of interest image data, and sign library data; A corresponding relationship is established between the associated data in the to-be-processed data whose positional relationship with each point of interest meets the set requirements and the identifier of each point of interest.

12. The method according to any one of claims 1 to 11, further comprising: In a case where the associated data includes public travel trajectory data, a denoising operation and / or a thinning operation is performed on the public travel trajectory data.

13. A guidance point acquisition device, comprising: A candidate set acquisition module, used to acquire a candidate set associated with the point of interest based on the associated data of the point of interest; wherein the candidate set includes a plurality of candidate points; A filtering module, configured to filter the plurality of candidate points in the candidate set based on the orientation area of ​​the point of interest; The guide point acquisition module is used to acquire the guide point associated with the interest point based on the candidate points retained after filtering.

14. The device according to claim 13, wherein: The candidate set acquisition module includes one or more of the following submodules: A delivery point submodule, configured to obtain a delivery point based on one or more of the public travel trajectory data, the driving trajectory data and the road network data in the associated data, and add the delivery point to a candidate set; A binding point submodule, configured to obtain a binding point based on one or more of the public travel trajectory data, the driving trajectory data and the road network data in the associated data, and add the binding point to a candidate set; The trajectory point submodule is used to add the trajectory points of the public travel trajectory data and / or the driving trajectory data in the associated data to a candidate set.

15. The device according to claim 13 or 14, wherein: The filtering module comprises: A calculation submodule, configured to calculate a direction area of ​​the point of interest based on the associated data; The processing submodule is used to remove the candidate points that are not in the direction area of ​​the point of interest from the candidate set to obtain the retained first candidate points.

16. The device according to claim 15, wherein: The calculation submodule is used to predict the orientation vector of the point of interest or the orientation fan area of ​​the point of interest based on one or more of the public travel trajectory data, driving trajectory data, point of interest image data, road network link data of the point of interest orientation, and sign library data in the associated data; wherein the orientation vector is used to determine the orientation triangle area of ​​the point of interest.

17. The device according to any one of claims 13 to 16, wherein: The guidance point acquisition module comprises: A sorting submodule, used to sort the plurality of first candidate points based on the features of the plurality of retained first candidate points to obtain a first sorting result; A selection submodule is used to select a first number of second candidate points from the first sorting results.

18. According to the device of claim 17, the characteristics of the first candidate point include one or more of the following: the straight-line distance to the point of interest, the walking distance to the point of interest, the ratio of the walking distance to the straight-line distance, the traffic volume of the road network link, the road network grade of the road network link, the traffic volume of public travel trajectory data and the traffic volume of driving trajectory data.

19. The method according to claim 18, wherein: The road network link includes a lane-level link in the road network, and the driving trajectory data includes lane-level driving trajectory data.

20. The device according to any one of claims 17 to 19, wherein: The guidance point acquisition module comprises: A clustering submodule, configured to perform clustering processing on the first number of second candidate points to obtain a clustering result, wherein the clustering result includes a third candidate point corresponding to each cluster center; A determination submodule is used to determine a guide point of the point of interest based on the third candidate point.

21. The device according to claim 20, wherein: The clustering submodule is used to cluster the first number of second candidate points based on the K nearest neighbor algorithm to obtain one or more clusters; and use the second candidate point closest to the cluster center of the one or more clusters as the third candidate point.

22. The device according to claim 20 or 21, wherein: The determination submodule is used to sort the multiple third candidate points to obtain a second sorting result; and select the third candidate point closest to the point of interest from the second sorting result as the guide point.

23. The apparatus according to any one of claims 13 to 22, further comprising: A data acquisition module, used to acquire data to be processed from multiple data sources; wherein the data to be processed includes one or more of the following: public travel trajectory data, driving trajectory data, road network data, point of interest orientation data, point of interest image data and sign library data; The association module is used to establish a corresponding relationship between the associated data whose positional relationship with each point of interest in the data to be processed meets the set requirements and the identifiers of each point of interest.

24. The apparatus according to any one of claims 13 to 23, further comprising: The data processing module is used for performing a denoising operation and / or a thinning operation on the public travel trajectory data when the associated data includes the public travel trajectory data.

25. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 12.

26. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-12.

27. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 12.

Images