Navigation route decision-making method, device, equipment, storage medium and product

Abstract

The invention discloses a navigation route decision-making method and device, equipment, a storage medium and a product, and relates to the technical field of vehicle navigation, the navigation route decision-making method comprises the following steps: in response to a navigation task instruction input by a user, extracting key information in the navigation task instruction; based on the key information, integrating vehicle historical travel data conforming to the key information in a preset vehicle end database into a candidate travel data set; and inputting the navigation task instruction and the candidate travel data set into the large model for matching verification, and determining a corresponding navigation route according to an output matching verification result. According to the method, the historical travel data is subjected to structured storage, when the user inputs the navigation task instruction, key information of different dimensions is extracted according to the navigation task instruction, the historical travel data conforming to the key information is integrated into the candidate travel data set, and then the most matched navigation route is judged and decided through a large model; and effective support for the memory-based navigation instruction is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle navigation, in particular to a navigation route decision method and device, equipment, storage medium and product. BACKGROUND

[0002] With the rapid evolution of vehicle navigation systems, more and more users are no longer satisfied with single-path navigation, but expect the system to remember the areas the vehicle has passed through, such as navigating to routes that have been traveled, scenic spots that have been passed through, and stores that have been visited. However, existing vehicle navigation systems have obvious deficiencies in handling complex tasks related to navigation memory, and are difficult to meet the increasingly diverse needs of users. SUMMARY

[0003] The main purpose of the present application is to provide a navigation route decision method, device, equipment, storage medium and product, aiming to solve the technical problem of the incompleteness of the existing vehicle navigation system in the storage of historical travel data, which leads to the inability to effectively support memory-based navigation instructions.

[0004] To achieve the above purpose, the present application provides a navigation route decision method, which comprises: In response to a navigation task instruction input by a user, extracting key information in the navigation task instruction, the key information including at least one of semantic dimension key information, time dimension key information and geographical dimension key information; Based on the key information, integrating vehicle historical travel data in a preset vehicle database that meets the key information into a candidate travel data set, wherein the vehicle historical travel data is obtained by structurally storing at least one of travel time information, effective driving trajectory, key passing point information and key passing point surrounding associated information; Inputting the navigation task instruction and the candidate travel data set into a preset large model for matching verification, and determining the corresponding navigation route according to the output matching verification result.

[0005] In an embodiment, the step of integrating vehicle historical travel data in a preset vehicle database that meets the key information into a candidate travel data set includes: Obtaining vehicle map navigation routes, vehicle actual driving trajectories and travel time information; Inputting the vehicle map navigation routes and the vehicle actual driving trajectories into a preset large model for path fitting and vehicle passing point data deduplication to obtain vehicle effective driving trajectories; Determining key passing point information in the vehicle effective driving trajectories and corresponding expanded key passing point surrounding associated information; The at least one of the travel time information, the effective driving track of the vehicle, the key point information and the key point surrounding associated information is structured as vehicle historical travel data and stored in a preset vehicle terminal database.

[0006] In an embodiment, the key point information includes category information of the key point, and after the step of structuring the at least one of the travel time information, the effective driving track of the vehicle, the key point information and the key point surrounding associated information as vehicle historical travel data and storing in a preset vehicle terminal database, the method further includes: performing semantic vector extraction on the category information and the key point surrounding associated information to obtain a semantic dimension vector index; updating the semantic dimension vector index to a preset local index library to match the semantic dimension key information.

[0007] In an embodiment, the method further includes, before the steps of obtaining the vehicle map navigation route, the actual driving track of the vehicle and the travel time information: when detecting that a user completes a boarding action, identifying the user according to an identity recognition result to determine a user identity code; querying whether the user identity code authorizes storage of vehicle historical travel data; if the vehicle historical travel data is authorized to be stored, establishing a binding relationship between the vehicle historical travel data stored in real time and the user identity code, and notifying the user that travel data collection and encrypted storage will be performed through voice broadcast.

[0008] In an embodiment, the method further includes, before the steps of obtaining the vehicle map navigation route, the actual driving track of the vehicle and the travel time information: obtaining a user identity code of a current user, and confirming whether the user identity code authorizes extraction of vehicle historical travel data of the current user from a preset vehicle terminal database; if the vehicle historical travel data is authorized, performing semantic dimension fuzzy matching on the key point information and the key point surrounding associated information of the vehicle historical travel data according to semantic dimension key information and in combination with a preset local index library to obtain first candidate travel data conforming to the semantic dimension; performing time dimension fuzzy matching on the travel time information of the vehicle historical travel data according to time dimension key information to obtain second candidate travel data conforming to the time dimension; performing geographical dimension fuzzy matching on the effective driving track of the vehicle historical travel data according to geographical dimension key information to obtain third candidate travel data conforming to the geographical dimension; aggregating the first candidate trip data, the second candidate trip data and the third candidate trip data into a temporary data set; de-duplicating the trip data in the temporary data set to obtain a de-duplicated temporary data set; based on the key information in different dimensions, assigning scores to the trip data in the de-duplicated temporary data set according to the score assignment results, and sorting the trip data in the de-duplicated temporary data set according to the score assignment results, to obtain a candidate trip data set sorted according to the correlation strength.

[0009] In an embodiment, the trip data includes a vehicle effective driving trajectory, category information of a key passing point, and surrounding associated information of a key passing point. The step of assigning scores to the trip data in the de-duplicated temporary data set according to the score assignment results, and sorting the trip data in the candidate trip data set according to the score assignment results, includes: assigning a first correlation strength score to the vehicle effective driving trajectory and determining a corresponding first weight to obtain a first score; assigning a second correlation strength score to the category information of the key passing point and determining a corresponding second weight to obtain a second score; assigning a third correlation strength score to the surrounding associated information of the key passing point and determining a corresponding second weight to obtain a third score; based on the first score, the second score and the third score, determining a score assignment result, and sorting the trip data in the candidate trip data set according to the score assignment result.

[0010] In an embodiment, the step of determining a corresponding navigation route according to the output matching verification result includes: if the matching verification result is that there is a key passing point that meets the navigation task instruction, the basic information of the key passing point is called, the key passing point is set as the destination, and the corresponding navigation route is determined; if the matching verification result is that there are two or more key passing points that meet the navigation task instruction, a list of candidate key passing points is confirmed, and the list of candidate key passing points is output to a preset terminal, so that the key passing point selected by the preset terminal based on the list of candidate key passing points is set as the destination, and the corresponding navigation route is determined; if the matching verification result is that there is no key passing point that meets the navigation task instruction, the result is announced through voice broadcast; and / or, the basic information of a real-time surrounding associated location is obtained, a list of real-time surrounding associated locations is constructed, and the list of candidate surrounding associated locations is output to a preset terminal, so that the surrounding associated location selected by the preset terminal based on the list of candidate surrounding associated locations is set as the destination, and the corresponding navigation route is determined.

[0011] In addition, to achieve the above object, the application further provides a navigation route decision device, which comprises: An extraction module is configured to extract key information in the navigation task instruction in response to a user input, wherein the key information comprises at least one of semantic dimension key information, time dimension key information and geographical dimension key information; A matching module is configured to integrate vehicle historical travel data in a preset vehicle end database that meets the key information into a candidate travel data set based on the key information, wherein the vehicle historical travel data is obtained by structurally storing at least one of travel time information, effective driving track, key passing point information and key passing point surrounding associated information; A verification module is configured to input the navigation task instruction and the candidate travel data set into a preset large model for matching verification, and determine a corresponding navigation route according to an output matching verification result.

[0012] In addition, to achieve the above object, the application further provides a navigation route decision device, which comprises:

[0013] In addition, to achieve the above object, the application further provides a storage medium, which is a computer readable storage medium, and a computer program is stored on the storage medium, wherein the computer program is executed by a processor to implement the steps of the navigation route decision method.

[0014] In addition, to achieve the above object, the application further provides a computer program product, which comprises a computer program, wherein the computer program is executed by a processor to implement the steps of the navigation route decision method.

[0015] The one or more technical solutions provided by the application have at least the following technical effects: Compared with the related art, recording start and end point positioning coordinates and rough time stamps, when processing complex tasks related to navigation memory, showing the entire vehicle historical driving route, waiting for the user to manually screen and confirm before executing the navigation task, resulting in low execution efficiency when processing memory-based navigation instructions, compared with the related art, the application extracts key information in the navigation task instruction in response to the user input navigation task instruction, the key information includes at least one of semantic dimension key information, time dimension key information and geographic dimension key information; Based on the key information, the vehicle historical travel data in the preset vehicle terminal database that meets the key information is integrated into a candidate travel data set, wherein the vehicle historical travel data is obtained by structurally storing at least one of travel time information, effective driving track, key passing point information and key passing point surrounding associated information; The navigation task instruction and the candidate travel data set are input into a preset large model for matching verification, and the corresponding navigation route is determined according to the output matching verification result. It can be understood that, by structurally storing the travel time information, the effective driving track, the key passing point information and the key passing point surrounding associated information, the vehicle historical travel data is obtained and pre-stored in the vehicle terminal database. When the user inputs the navigation task instruction, the application extracts key information of different dimensions according to the navigation task instruction, matches the structured vehicle historical travel data in the vehicle terminal database, integrates the vehicle historical travel data meeting the time dimension or the geographic dimension or the semantic dimension into the candidate travel data set, and then judges the vehicle historical travel data in the candidate travel data set that matches the navigation task instruction through the large model to generate the corresponding navigation route, thereby realizing effective support for memory-based navigation instructions. BRIEF DESCRIPTION OF DRAWINGS

[0016] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present application and, together with the specification, serve to explain the principles of the application.

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0018] Figure 1 The flowchart provided by the navigation route decision method embodiment one of the present application; Figure 2 The flowchart provided by the navigation route decision method embodiment two of the present application; Figure 3 The brief flowchart of the navigation route decision method provided by the embodiment two of the present application; Figure 4 A schematic diagram of a module structure of a navigation route decision device according to an embodiment of the present application; Figure 5 A schematic diagram of a device structure of a hardware operating environment involved in a navigation route decision method according to an embodiment of the present application.

[0019] The object implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0020] It should be understood that the specific embodiments described herein are merely intended to explain the technical solutions of the present application, and are not intended to limit the present application.

[0021] In order to better understand the technical solutions of the present application, the specific embodiments will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] The main solution of the embodiment of the present application is: In response to a navigation task instruction input by a user, key information in the navigation task instruction is extracted, the key information including at least one of semantic dimension key information, time dimension key information and geographic dimension key information; Based on the key information, vehicle historical travel data in a preset vehicle-side database that meets the key information is integrated into a candidate travel data set, wherein the vehicle historical travel data is obtained by structurally storing at least one of travel time information, effective travel trajectory, key passing point information and key passing point surrounding associated information; The navigation task instruction and the candidate travel data set are input into a preset large model for matching verification, and a corresponding navigation route is determined according to the output matching verification result.

[0023] In the present embodiment, the navigation route decision device is taken as the execution subject, and for the sake of brevity, the device will be described in detail below.

[0024] With the rapid evolution of vehicle-side navigation systems, more and more users no longer satisfy with single-path navigation, but expect the system to remember the areas the vehicle has passed through, such as navigating to the routes once traveled, passing through the scenic spots, and stopping at the shops. However, the existing vehicle-side navigation has obvious deficiencies in handling complex tasks related to navigation memory, and it is difficult to meet the increasingly diverse scene needs of users.

[0025] Since the existing technology only records the start and end point positioning coordinates and rough time stamps to save the vehicle historical travel route, when handling complex tasks related to navigation memory, all the vehicle historical travel routes are displayed, and the user needs to manually screen and confirm before the navigation task can be executed, resulting in low efficiency in handling navigation instructions based on memory.

[0026] The application provides a solution. By structurally storing travel time information, effective travel trajectory, key pass point information and key pass point surrounding associated information, vehicle historical travel data is obtained and pre-stored in a vehicle terminal database. When a user inputs a navigation task instruction, the application extracts key information of different dimensions according to the navigation task instruction, matches the vehicle historical travel data stored in the vehicle terminal database in a structured manner, integrates the vehicle historical travel data meeting the time dimension or the geographic dimension or the semantic dimension into a candidate travel data set, and then judges the vehicle historical travel data in the candidate travel data set that matches the navigation task instruction through a large model to generate a corresponding navigation route, thereby effectively supporting the navigation instruction based on memory.

[0027] Based on this, the navigation route decision method provided in the embodiments of the application is provided with reference to Figure 1 , Figure 1 The flowchart of the first embodiment of the navigation route decision method of the application is shown in the figure.

[0028] In this embodiment, the navigation route decision method includes steps S10-S30: Step S10, in response to a navigation task instruction input by a user, key information in the navigation task instruction is extracted, the key information including at least one of semantic dimension key information, time dimension key information and geographic dimension key information; It should be noted that the navigation task instruction input by the user refers to a natural language instruction related to navigation issued by the user through voice or text, such as navigation to a place visited or query of a specific place in the historical travel. The extraction of key information refers to the analysis and recognition of the navigation task instruction through natural language processing technology to separate the key elements in the instruction, including but not limited to semantic, time and geographic information. When extracting key information, the context chat record in the current active conversation can be referred to. The semantic dimension key information refers to the place type, use category or scene description involved in the instruction, such as entity concepts such as coffee shop or restaurant. The time dimension key information refers to the time-related description contained in the instruction, such as the time range limit of last week or the past month. The geographic dimension key information refers to the geographic position or area description mentioned in the instruction, such as urban area or lake.

[0029] It can be understood that this step realizes fine-grained analysis of complex natural language instructions by extracting multi-dimensional key information from the navigation task instruction input by the user. This process decomposes the user's intention into independent feature dimensions such as semantics, time, and geography, providing clear input for subsequent matching retrieval based on structured historical data. The technical effect is to improve the system's understanding of ambiguous or complex queries, enabling the navigation system to more accurately identify user needs and support tasks such as navigating to historical waypoints or querying locations within a specific spatio-temporal range, thereby enhancing the adaptability and response efficiency of the navigation memory function.

[0030] In step S20, based on the key information, the vehicle historical travel data in the preset vehicle-side database that meets the key information is integrated into a candidate travel data set, wherein the vehicle historical travel data is obtained by structuring at least one of travel time information, effective travel trajectory, key waypoint information, and key waypoint surrounding associated information; It should be noted that the preset vehicle-side database refers to an encrypted database stored locally by the vehicle, used to save structured vehicle historical travel data. The key information that meets the key information means that the travel records consistent with each dimension of the key information are filtered out by a matching algorithm. The candidate travel data set refers to a temporary travel set formed after multi-path recall and preliminary screening. The travel time information includes the start time stamp and the end time stamp. The effective travel trajectory refers to an optimized path sequence after removing abnormal points and repeated points through big model analysis and processing. The key waypoint information includes the POI (point of interest) basic information and its refined category annotation of the location point and the route turning point that stays more than a preset time. The key waypoint surrounding associated information refers to the relevant location information within a preset radius around the key waypoint.

[0031] It can be understood that this step realizes efficient screening of structured historical travel data in the vehicle-side database through a multi-dimensional parallel retrieval mechanism. The technical effect is to make full use of the multi-dimensional data storage system established in the early stage, and through the coordinated matching of the three independent dimensions of semantics, time, and geography, it can effectively deal with the ambiguity and incompleteness that may exist in user instructions. This multi-path recall mechanism significantly improves the recall rate of relevant historical travel, especially for complex queries that integrate spatio-temporal and semantic features, ensuring as many relevant travels as possible into the candidate set, providing a sufficient data foundation for subsequent big model accurate decision-making, thereby improving the accuracy and completeness of the system in handling navigation memory tasks.

[0032] In step S30, the navigation task instruction and the candidate travel data set are input into a preset big model for matching verification, and the corresponding navigation route is determined according to the output matching verification result.

[0033] It should be noted that the preset large model refers to a language model that is pre-trained and optimized for navigation scenarios, and has the ability of semantic understanding, logical reasoning and decision-making. The matching verification refers to the process of semantic consistency analysis and context relevance verification of the large model on the navigation task instruction and the candidate trip data set. The matching verification result refers to the decision conclusion output by the large model after analysis, including single matching result, multiple matching results or no matching result, etc. Determining the corresponding navigation route refers to generating a specific executable navigation path plan according to the matching verification result, including destination coordinate setting and route guidance scheme.

[0034] It can be understood that this step realizes intelligent filtering and decision optimization of candidate trip data by introducing a large model for final matching verification. The technical effect is to use the powerful semantic understanding and reasoning ability of the large model to accurately verify the context of the candidate results obtained by multi-dimensional retrieval, effectively solving the ambiguity problem that may be caused by simply based on keyword matching. This mechanism can identify the target location that best meets the user's needs from multiple similar candidate trips according to the real intention of the user's instruction, and provide reasonable alternative solutions when there is no direct match, thereby significantly improving the decision accuracy and user experience of the navigation memory system, and ensuring the smoothness and intelligence of the navigation task execution.

[0035] The embodiment provides a navigation route decision method, which stores vehicle historical trip data by structuring storage of trip time information, effective driving track, key passing point information and key passing point surrounding associated information, and pre-stores the vehicle historical trip data in a vehicle-side database. When a user inputs a navigation task instruction, the application extracts key information of different dimensions according to the navigation task instruction, matches the structured storage of vehicle historical trip data in the vehicle-side database, integrates vehicle historical trip data meeting time dimension or geographical dimension or semantic dimension into a candidate trip data set, and then judges vehicle historical trip data in the candidate trip data set that matches the navigation task instruction through a large model to generate a corresponding navigation route, thereby realizing effective support for navigation instructions based on memory.

[0036] Based on the first embodiment of the application, in the second embodiment of the application, the same or similar contents as the above embodiment one can refer to the above introduction, and the subsequent will not be repeated. The step of integrating vehicle historical trip data in the preset vehicle-side database that meets the key information into a candidate trip data set includes: Obtaining vehicle map navigation route, vehicle actual driving track and trip time information; Inputting the vehicle map navigation route and the vehicle actual driving track to a preset large model for path fitting and vehicle passing point data deduplication to obtain vehicle effective driving track; determine the key passing point information in the effective driving track of the vehicle and the corresponding expanded key passing point surrounding associated information; structurally store at least one of the travel time information, the effective driving track of the vehicle, the key passing point information and the key passing point surrounding associated information as vehicle historical travel data and save to a preset vehicle-side database.

[0037] It should be noted that obtaining the vehicle map navigation route refers to the planning route data obtained in real time from the vehicle navigation system, including path point sequence and passing point information. The actual driving track of the vehicle refers to the complete path sequence formed by connecting the starting position data, the ending position data and the vehicle passing point data. The travel time information is the time range data calculated based on the travel starting time and the travel ending time. Path fitting refers to the process of fusing and analyzing the vehicle map navigation route and the actual driving track of the vehicle through a large model to generate an optimized path. The effective driving track of the vehicle refers to the optimized track obtained after path fitting and de-duplication processing, which excludes positioning errors and redundant path points. The key passing point surrounding associated information refers to the surrounding facility information collected within a preset expansion radius range and spatially associated with the key passing point. The vehicle historical travel data refers to a data set containing at least one of the travel time information, the effective driving track of the vehicle, the key passing point information and the key passing point surrounding associated information. The preset vehicle-side database refers to an embedded database stored locally by the vehicle, used to safely save historical travel data.

[0038] It can be understood that the embodiment establishes a complete data preprocessing and storage mechanism, providing a high-quality data basis for subsequent multi-dimensional fuzzy recall. The technical effect is that through the fusion analysis of the original navigation route and the actual driving track, combined with the intelligent processing capability of the large model, the accuracy and completeness of the track data are effectively improved. Through the identification of the key passing point and the expansion of the surrounding information, the semantic dimension and context information of the travel data are enriched, creating conditions for subsequent fuzzy retrieval based on multiple attributes such as semantics, time and geography. This structured data storage method not only solves the technical problem of incomplete storage of historical data of existing navigation systems, but also provides sufficient data support for efficient processing of navigation instructions based on memory, significantly improving the overall performance and user experience of the navigation memory system.

[0039] Exemplarily, the device acquires a trip start timestamp and a start location data of the vehicle in response to a vehicle trip start signal, collects vehicle passing point data according to a preset time interval until a trip end timestamp and an end location data of the vehicle are acquired in response to a vehicle trip end signal; the device determines an actual driving track of the vehicle according to the start location data, the end location data and the vehicle passing point data; the device determines trip time information according to the trip start timestamp and the trip end timestamp; the device acquires a vehicle map navigation route, inputs the vehicle map navigation route and the actual driving track of the vehicle into a preset large model for path fitting, and removes the vehicle passing point data in the actual driving track of the vehicle based on a path fitting result through the large model to obtain an effective driving track of the vehicle; the device determines a key passing point in the effective driving track of the vehicle according to a preset key passing point screening rule, and labels actual navigation information and use category information of the key passing point as key passing point information; the device collects associated location information of the key passing point according to a preset expansion radius, and integrates the collected associated location information as key passing point surrounding associated information; the device stores the trip time information, the effective driving track of the vehicle, the key passing point information and the key passing point surrounding associated information as vehicle historical trip data in a structured manner, encrypts and saves the vehicle historical trip data to a preset vehicle end database.

[0040] It should be noted that the vehicle trip start signal refers to an event signal for triggering trip start recording, including but not limited to vehicle ignition start, user manual start of trip recording or system automatic detection of vehicle movement, etc. The trip start timestamp refers to accurate time information recorded at the start of the trip, in a standard time format. The start location data refers to the starting point longitude and latitude coordinates of the trip acquired through the vehicle-mounted positioning module. The preset time interval refers to the positioning data collection period set by the system. The vehicle passing point data refers to the serialized location coordinate points collected continuously at time intervals during the trip. The vehicle trip end signal refers to an event signal for triggering trip end recording, including but not limited to vehicle engine stop, user manual end of trip or system detection of long-term vehicle stationary, etc. The trip end timestamp refers to accurate time information recorded at the end of the trip. The end location data refers to the longitude and latitude coordinates of the vehicle at the end of the trip.

[0041] Additionally, it is necessary to point out that the actual driving trajectory of the vehicle refers to the complete path sequence formed by connecting the starting position data, the ending position data, and the vehicle waypoint data. The travel time information refers to the travel time range information calculated based on the travel start timestamp and the travel end timestamp, including the start time, the end time, and the duration. The vehicle map navigation route refers to the planned route data obtained from the vehicle navigation system, including but not limited to the path point sequence and the waypoint information. The pre-set large model refers to a language model that has been pre-trained and has path analysis capabilities, used to process trajectory data. Path fitting refers to the process of fusing and analyzing the vehicle map navigation route and the actual driving trajectory of the vehicle through the large model to generate an optimized path. The effective driving trajectory of the vehicle refers to the optimized trajectory obtained after path fitting and de-duplication processing, excluding positioning errors and redundant path points.

[0042] Additionally, it is necessary to point out that the key waypoint selection rule refers to a set of conditions pre-set by the system to identify important location points in the trajectory, including but not limited to the stay time threshold, the path turning angle threshold, etc. The key waypoint refers to an important location point extracted from the effective driving trajectory of the vehicle according to the selection rule, such as a long-stay point or a path turning point. The actual navigation information refers to the basic data of the key waypoint obtained from the map service API (Application Programming Interface, application programming interface), including but not limited to the place name, the latitude and longitude coordinates, and the classification label. The use category information refers to the refined classification description of the functional attributes of the key waypoint, such as coffee shops, parking lots, or scenic spots, etc. The pre-set expansion radius refers to the system-set search range of the surrounding places. The associated place information refers to the surrounding facility information that has a spatial association with the key waypoint within the expansion radius. The key waypoint surrounding association information refers to the structured data set formed by integrating all the associated place information. Structured storage refers to the process of organizing and managing different categories of travel data in a unified format. The vehicle historical travel data refers to the complete data set containing travel time information, effective driving trajectory of the vehicle, key waypoint information, and key waypoint surrounding association information. The pre-set vehicle-side database refers to the embedded database stored locally in the vehicle, used to safely save the encrypted historical travel data.

[0043] It can be understood that the embodiment establishes a high-quality data foundation for the navigation memory function by constructing a complete travel data collection, cleaning and storage process. The technical effect is to realize intelligent optimization processing of the original driving trajectory, effectively eliminate invalid data such as positioning error, path fluctuation caused by traffic congestion and short distance movement through the path fitting and deduplication capability of the large model, and improve the accuracy and availability of the trajectory data. At the same time, through the deep mining of the key passing points and the expansion of the surrounding information, the semantic dimension and the context information of the travel data are enriched, so that the stored historical data not only contains accurate space-time trajectory, but also integrates rich scene semantic description. This multi-dimensional structured storage mechanism provides sufficient data support for subsequent fuzzy retrieval based on multiple attributes such as semantics, time and geography, and fundamentally solves the technical bottleneck of incomplete historical data storage and POI (point of interest) information missing of the existing navigation system, and lays a solid foundation for efficient processing of memory-based navigation instructions.

[0044] In a feasible embodiment, the key passing point information includes category information of the key passing point, and after the step of structuring and storing at least one of the travel time information, the effective driving trajectory of the vehicle, the key passing point information and the key passing point surrounding associated information as vehicle historical travel data and saving to a preset vehicle-side database, the method further includes: extracting semantic vectors from the category information and the key passing point surrounding associated information to obtain a semantic dimension vector index; updating the semantic dimension vector index to a preset local index library to match the semantic dimension key information.

[0045] It should be noted that semantic vector extraction refers to a calculation process of converting the use category information and the surrounding associated information in text format into numerical vector representation through a natural language processing model. The semantic dimension vector index refers to an efficient retrieval data structure established by these semantic vectors. The preset local index library refers to a local database component in the vehicle-side embedded system specially used for storing and managing semantic vectors. Matching the semantic dimension key information refers to a process of similarity calculation and retrieval of semantic content in the user query by using the constructed vector index.

[0046] It can be understood that the embodiment realizes efficient semantic matching capability of unstructured text information by constructing a semantic dimension vector index. The technical effect is that the fuzzy semantic expression in the user natural language query is deeply associated with the place category information in the historical trip, and the over-reliance of the traditional keyword matching method on consistent expression is overcome. The vector-based semantic retrieval mechanism can effectively support the understanding and matching of synonyms, near-synonyms and associated scenes, significantly improve the response accuracy and recall rate of the system to fuzzy semantic queries, and provide core technical support for realizing intelligent navigation memory retrieval. At the same time, through the maintenance of the local index library, the retrieval efficiency is ensured while the user data privacy security is ensured.

[0047] In a feasible embodiment, the steps of acquiring the vehicle map navigation route, the actual driving trajectory of the vehicle, and the travel time information include the following steps before the steps: When it is detected that the user completes the getting-on action, the identity of the user is identified, and the user identity code is determined according to the identification result; Query whether the user identity code authorizes storage of vehicle historical travel data; If the storage of vehicle historical travel data is authorized, a binding relationship between the current real-time stored vehicle historical travel data and the user identity code is established, and the user is informed by voice broadcast that the travel data collection and encrypted storage will be performed.

[0048] It should be noted that detecting that the user completes the getting-on action means that the user enters the vehicle and sits down through the vehicle sensor or camera recognition trigger event. The identity of the user is identified, which means that the user's identity is confirmed by using face recognition technology or other biometric verification methods. The user identity code is a unique user identification string generated after identity recognition, which is used to distinguish different users. Querying whether the user identity code authorizes storage of vehicle historical travel data means accessing a local authorization list to verify whether the user has agreed to collect and store historical data. The binding relationship is established, which means that the real-time collected travel data is associated with the user identity code, and the data ownership is clear. The voice broadcast notification means that the voice prompt information is output to the user through the vehicle audio system.

[0049] It can be understood that the embodiment realizes personalized management and privacy protection of user data through the pre-identity recognition and authorization verification mechanism. The technical effect is to ensure the legality and pertinence of historical travel data collection and storage, and to avoid data confusion between multiple users. By establishing a binding relationship between the user identity code and the travel data, an accurate data basis is provided for subsequent personalized navigation memory services. At the same time, voice broadcast notification enhances system transparency, allowing users to clearly understand data collection status, improving user trust and experience satisfaction of the system. This mechanism provides a feasible solution for navigation memory function in a multi-user shared vehicle scenario under the premise of ensuring data security.

[0050] Exemplarily, with reference to Figure 2 When the system detects that the user completes the boarding action, the face recognition module is automatically started for identity authentication, and a unique user identity code is generated. The system then queries whether the user identity code has authorized historical route storage function. If the user has not authorized, the system only provides basic navigation service and does not collect or store any historical travel data. If the user has authorized, the system will establish a binding relationship between the current travel data and the user identity code, and at the same time, through voice broadcast, it will explicitly prompt the user that the personalized navigation memory function has been turned on, and inform that all data is only stored in the local machine after encryption to fully protect user privacy. This process realizes effective isolation of multi-user data, avoids information confusion, and at the same time, improves personalized experience while strengthening privacy and security protection.

[0051] In a feasible implementation, the step of integrating the vehicle historical travel data in the preset vehicle-side database that meets the key information into a candidate travel data set based on the key information includes: Obtaining a user identity code of a current user, and confirming whether the user identity code authorizes extraction of vehicle historical travel data of the current user from a preset vehicle-side database; If authorized, according to semantic dimension key information, combining a preset local index library, performing semantic dimension fuzzy matching on the key passing point information and the key passing point surrounding associated information of the vehicle historical travel data to obtain first candidate travel data meeting the semantic dimension; According to time dimension key information, performing time dimension fuzzy matching on the travel time information of the vehicle historical travel data to obtain second candidate travel data meeting the time dimension; According to geographic dimension key information, performing geographic dimension fuzzy matching on the effective travel trajectory of the vehicle historical travel data to obtain third candidate travel data meeting the geographic dimension; Summarizing the first candidate travel data, the second candidate travel data and the third candidate travel data into a temporary data set; de-duplicate the travel data in the temporary data set to obtain a de-duplicated temporary data set; based on the key information in different dimensions, assigning a correlation strength score to the travel data in the de-duplicated temporary data set, and sorting the travel data in the de-duplicated temporary data set according to the score result to obtain a candidate travel data set sorted by correlation strength.

[0052] It should be noted that obtaining the user identity code of the current user refers to extracting the authenticated user unique identifier from the current active session. Confirming the authorization state refers to verifying whether the user identity code has the permission to read the historical travel data from the vehicle end database. The fuzzy matching of semantic dimensions refers to using the semantic vector index in the local index library to perform vector similarity calculation on the key passing point information and the surrounding associated information. The first candidate travel data refers to the preliminary screening result set obtained through semantic matching. The fuzzy matching of time dimensions refers to range comparison of the time key information in the user instruction and the travel time information. The second candidate travel data refers to the preliminary screening result set obtained through time matching. The fuzzy matching of geographic dimensions refers to spatial relationship analysis of the geographic key information in the user instruction and the effective travel trajectory. The third candidate travel data refers to the preliminary screening result set obtained through geographic matching. The correlation strength score refers to the quantitative evaluation of the relevance of the travel and the query instruction through multi-dimensional score weighted summation.

[0053] It can be understood that the embodiment realizes comprehensive coverage and accurate response to the user query intention by establishing a multi-dimensional parallel retrieval and fusion sorting mechanism. The technical effect is to fully utilize the structured data system and semantic index constructed in the early stage, and through parallel retrieval of the semantic, time, and geographic three independent channels, the recall rate of the relevant historical travel is significantly improved. Especially through the multi-path recall strategy, it can effectively capture various clues that may be implied in the user's fuzzy query, and avoid missing relevant results due to insufficient single-dimensional matching. The subsequent weighted scoring and sorting mechanism comprehensively considers the contribution of each dimension evidence, ensuring that the final output candidate travel set is comprehensive and orderly, providing a high-quality data basis for the accurate decision of the subsequent large model, thereby improving the query success rate and user satisfaction of the navigation memory system as a whole.

[0054] In a feasible embodiment, the travel data includes vehicle effective travel trajectory, category information of key passing points, and key passing point surrounding associated information. The step of assigning a correlation strength score to the travel data in the de-duplicated temporary data set, and sorting the travel data in the candidate travel data set according to the score result, comprises: The first correlation strength score is assigned to the effective driving track of the vehicle, and a corresponding first weight is determined to obtain a first score; The second correlation strength score is assigned to the category information of the key passing point, and a corresponding second weight is determined to obtain a second score; The third correlation strength score is assigned to the surrounding associated information of the key passing point, and a corresponding second weight is determined to obtain a third score; Based on the first score, the second score, and the third score, a score result is determined, and the trip data in the candidate trip data set is sorted according to the score result.

[0055] It should be noted that the correlation strength score refers to the process of evaluating the relevance of trip data and user query instructions through quantitative calculation. The first correlation strength score is assigned to the effective driving track of the vehicle, which means that the score is based on the matching degree of the driving track and the geographical dimension key information. The first weight refers to the weight coefficient assigned to the driving track matching score, which reflects the importance of this dimension in the overall evaluation. The first score is the final score of the driving track matching degree after weight adjustment.

[0056] The second correlation strength score is assigned to the category information of the key passing point, which means that the score is based on the matching degree of the category information and the semantic dimension key information. The second weight refers to the weight coefficient assigned to the category information matching score. The second score is the final score of the category information matching degree after weight adjustment. The third correlation strength score is assigned to the surrounding associated information of the key passing point, which means that the score is based on the matching degree of the surrounding associated information and the semantic dimension key information.

[0057] The third weight refers to the weight coefficient assigned to the surrounding associated information matching score. The third score is the final score of the surrounding associated information matching degree after weight adjustment. The score result refers to the total score value obtained by weighted summation. The sorting refers to the operation of arranging the trip records in the candidate trip data set in descending order according to the score result.

[0058] It can be understood that the present embodiment realizes fine evaluation and intelligent sorting of candidate trip data by establishing a multi-dimensional weighted scoring mechanism. The technical effect is that by assigning different weights to the matching results of different dimensions, the relative importance of each dimension evidence in the overall decision is accurately reflected. This multi-dimensional weighted evaluation method can more comprehensively consider the correlation degree of trip data and user query, avoiding the deviation that may be caused by single dimension matching. Through scientific weight allocation and quantitative scoring, the trip data most relevant to the user query intention can be presented first, providing clear and orderly input for subsequent large model decision-making, significantly improving the retrieval accuracy and decision reliability of the navigation memory system, thereby effectively improving the user experience when processing navigation instructions based on memory.

[0059] Exemplarily, after deduplication, the remaining routes are sorted in descending order of the association strength with the key information, and the association strength is calculated by the score rule of "multi-dimensional score weighted summation".

[0060] Score rule: route trajectory matching score (weight 0.3) + key POI category matching score (weight 0.5) + surrounding place matching score (weight 0.2).

[0061] The final output is a list of routes sorted by association strength, providing clear and ordered candidate data for subsequent large model decision-making.

[0062] In a possible implementation, the step of determining the corresponding navigation route according to the matching verification result of the output comprises: If the matching verification result is that there is one key pass point meeting the navigation task instruction, the basic information of the key pass point is called, the key pass point is set as the destination, and the corresponding navigation route is determined; If the matching verification result is that there are two or more key pass points meeting the navigation task instruction, a list of candidate key pass points is confirmed, the list of candidate key pass points is output to a preset terminal to wait for user selection, and after receiving a user selection instruction, the key pass point selected by the user is set as the destination, and the corresponding navigation route is determined; If the matching verification result is that there is no key pass point meeting the navigation task instruction, the announcement is made through voice broadcast, the basic information of the real-time surrounding associated place is obtained, the list of real-time surrounding associated places is constructed, the list of candidate surrounding associated places is output to a preset terminal to wait for user selection, and after receiving a user selection instruction, the surrounding associated place selected by the user is set as the destination, and the corresponding navigation route is determined.

[0063] It should be noted that the matching verification result refers to the judgment conclusion output by the large model after analyzing the candidate travel data set. The key way point meeting the navigation task instruction refers to the historical travel way point that is highly matched with the user navigation task instruction in the semantic, time and geographical dimensions. The basic information includes but is not limited to the name, latitude and longitude coordinates, address and other navigation necessary data of the key way point. Set as the destination refers to taking the key way point as the end point of the navigation path planning. The list of candidate key way points refers to the selection list formed by sorting the matched candidate way points according to the correlation strength. The preset terminal refers to the human-computer interaction interface such as the vehicle-mounted central control display screen or mobile device application program. The user selection instruction refers to the input operation of the user confirming from the candidate list through touch or voice. The voice broadcast refers to the voice notification to the user that there is no matching result through text-to-speech technology. The real-time surrounding associated location refers to the real-time available location information obtained by the map service within the preset range with the current position of the vehicle as the center.

[0064] It can be understood that the embodiment establishes a multi-scene adaptive decision mechanism to realize the complete task closed loop of the navigation memory system. The technical effect is to provide differentiated processing strategies according to different matching results, which not only ensures the execution efficiency in a single matching scene, but also solves the decision ambiguity problem in the multi-candidate situation through the interactive confirmation mechanism. Especially when there is no historical matching result, the system can intelligently switch to the real-time surrounding recommendation mode, effectively avoiding task interruption and maintaining the continuity of navigation service. This hierarchical processing mechanism significantly improves the adaptability and robustness of the system in actual use, ensuring that users can obtain a smooth and natural navigation experience in various scenes, and through clear human-computer interaction design, the understanding and control of the system decision process by the user is enhanced.

[0065] Exemplarily, in order to assist in understanding the implementation process of the navigation route decision method obtained after the above embodiment one, please refer to Figure 3 , Figure 3 A brief flowchart of a navigation route decision method is provided, specifically: First, the device responds to the navigation task instruction initiated by the user, and enters the vehicle-end intelligent agent processing stage. This stage corresponds to the extraction step in the claim, and the semantic dimension, time dimension and geographical dimension key information contained in the instruction are extracted through natural language processing technology.

[0066] Then, the device enters a decision-making stage, corresponding to the verification step of the claim. The device inputs the navigation task instruction and the candidate trip data set into the large model for matching verification. The large model outputs the matching verification result based on semantic understanding and context analysis.

[0067] Then, the device enters a decision-making stage, corresponding to the verification step of the claim. The device inputs the navigation task instruction and the candidate trip data set into the large model for matching verification. The large model outputs the matching verification result based on semantic understanding and context analysis.

[0068] According to different situations of the matching verification result, the device performs corresponding navigation route determination operations. When the number of candidates is zero, the device informs the user through voice broadcast that there is no matching result, and provides real-time surrounding location recommendation as an alternative solution. When there is only one matching result, the device directly sets the key passing point as the destination and generates a navigation route. When there are multiple candidates, the device outputs a candidate list and waits for user selection through a human-computer interaction interface, and finally determines the navigation route according to the user selection instruction.

[0069] The entire process embodies a complete technical solution from instruction analysis, data retrieval, intelligent verification to route determination. Through the synergistic effect of multi-dimensional fuzzy recall and large model decision, the execution efficiency and accuracy of the navigation memory task are effectively improved.

[0070] It should be noted that the above examples are only used to understand the present application and do not limit the navigation route decision method of the present application. Based on this technical concept, more forms of simple transformation are within the protection scope of the present application.

[0071] The present application also provides a navigation route decision device, please refer to Figure 4 The navigation route decision device comprises: An extraction module 10 is configured to extract key information in a navigation task instruction input by a user in response to the navigation task instruction, wherein the key information comprises at least one of semantic dimension key information, time dimension key information, and geographic dimension key information. A matching module 20 is configured to integrate vehicle historical trip data in a pre-set vehicle-side database that meets the key information into a candidate trip data set based on the key information, wherein the vehicle historical trip data is obtained by structurally storing at least one of trip time information, effective driving track, key passing point information, and key passing point peripheral association information. The verification module 30 is configured to input the navigation task instruction and the candidate route data set into a preset large model for matching verification, and determine a corresponding navigation route according to an output matching verification result.

[0072] And / or, the navigation route decision device comprises: The first acquisition module is configured to acquire a vehicle map navigation route, a vehicle actual driving track, and travel time information. The first de-duplication module is configured to input the vehicle map navigation route and the vehicle actual driving track into a preset large model for path fitting and de-duplication of vehicle passing point data, to obtain a vehicle effective driving track. The first determination module is configured to determine key passing point information in the vehicle effective driving track and corresponding expanded key passing point peripheral associated information. The first storage module is configured to structure at least one of the travel time information, the vehicle effective driving track, the key passing point information, and the key passing point peripheral associated information, and save the structured vehicle historical travel data to a preset vehicle terminal database.

[0073] And / or, the navigation route decision device comprises: The first extraction module is configured to perform semantic vector extraction on the use category information of the key passing point and the key passing point peripheral associated information, to obtain a semantic dimension vector index. The first update module is configured to update the semantic dimension vector index to a preset local index library, so as to match the semantic dimension key information.

[0074] And / or, the navigation route decision device comprises: The first identification module is configured to identify the user when detecting that the user completes a boarding action, and determine a user identity code according to an identification result. The first query module is configured to query whether the user identity code authorizes storage of vehicle historical travel data. The first binding module is configured to, if the storage of the vehicle historical travel data is authorized, establish a binding relationship between the vehicle historical travel data currently stored in real time and the user identity code, and notify the user that travel data collection and encrypted storage will be performed through voice broadcast.

[0075] And / or, the navigation route decision device comprises: The second extraction module is configured to acquire a user identity code of a current user, and confirm whether the user identity code authorizes extraction of vehicle historical travel data of the current user from a preset vehicle terminal database. The first matching module is configured to, if authorized, perform fuzzy matching of the key passing point information and the key passing point peripheral associated information of the vehicle historical travel data in a semantic dimension according to semantic dimension key information and in combination with a preset local index library, to obtain first candidate travel data conforming to the semantic dimension. The second matching module is configured to perform fuzzy matching of the travel time information of the vehicle historical travel data in a time dimension according to time dimension key information, to obtain second candidate travel data conforming to the time dimension. The third matching module is configured to perform fuzzy matching of the effective driving track of the vehicle historical travel data in a geographic dimension according to geographic dimension key information, to obtain third candidate travel data conforming to the geographic dimension. The first summarizing module is configured to summarize the first candidate travel data, the second candidate travel data and the third candidate travel data into a temporary data set. The second deduplication module is configured to deduplicate the travel data in the temporary data set, to obtain a deduplicated temporary data set. The first sorting module is configured to assign scores to the travel data in the deduplicated temporary data set based on the key information in different dimensions, and sort the travel data in the deduplicated temporary data set according to the score assignment result, to obtain a candidate travel data set sorted by correlation strength.

[0076] And / or, the first sorting module includes: The first scoring module is configured to assign a first correlation strength score to the effective driving track of the vehicle and determine a corresponding first weight, to obtain a first score. The second scoring module is configured to assign a second correlation strength score to the category information of the key passing point and determine a corresponding second weight, to obtain a second score. The third scoring module is configured to assign a third correlation strength score to the key passing point peripheral associated information and determine a corresponding second weight, to obtain a third score. The second sorting module is configured to determine a score assignment result based on the first score, the second score and the third score, and sort the travel data in the candidate travel data set according to the score assignment result.

[0077] And / or, the verification module 30 includes: The first navigation module is configured to, if the matching verification result is that there is a key passing point conforming to a navigation task instruction, call the basic information of the key passing point, set the key passing point as a destination, and determine a corresponding navigation route. The second navigation module is configured to, if the matching verification result is that there are two or more key pass points that match the navigation task instruction, confirm a list of candidate key pass points, output the list of candidate key pass points to a preset terminal to wait for a user to select, and set the key pass point selected by the user as a destination and determine a corresponding navigation route after receiving a user selection instruction. The third navigation module is configured to, if the matching verification result is that there is no key pass point that matches the navigation task instruction, report through voice broadcast, acquire basic information of real-time surrounding associated locations, construct a list of candidate surrounding associated locations, output the list of candidate surrounding associated locations to a preset terminal to wait for a user to select, and set the surrounding associated location selected by the user as a destination and determine a corresponding navigation route after receiving a user selection instruction.

[0078] The navigation route decision device provided in the present application adopts the navigation route decision method in the above embodiments, and can solve the technical problem of the incomplete history travel data storage of the existing vehicle terminal navigation system, which leads to the inability to effectively support the memory-based navigation instruction. Compared with the prior art, the navigation route decision device provided in the present application has the same beneficial effects as the navigation route decision method provided in the above embodiments, and other technical features in the navigation route decision device are the same as the features disclosed in the above embodiments, which will not be repeated here.

[0079] The present application provides a navigation route decision device, which comprises at least one processor and a memory connected with the at least one processor, wherein the memory stores instructions executable 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 the navigation route decision method in the above embodiment one.

[0080] Reference will be made to the following description of the embodiments of the present application with reference to the accompanying drawings. Figure 5 which shows a structural schematic diagram of a navigation route decision device suitable for implementing the embodiments of the present application. The navigation route decision device in the embodiments of the present application can include but is not limited to mobile terminals such as mobile phones, tablet computers, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PMPs (Portable Media Player), vehicle terminal (for example, vehicle navigation terminal) and the like, and fixed terminals such as digital televisions, desktop computers and the like. Figure 5 The navigation route decision device shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0081] As Figure 5As shown, the navigation route decision device can include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to programs stored in a read only memory (ROM) 1002 or loaded from a storage device 1003 into a random access memory (RAM) 1004. Various programs and data required for operation of the navigation route decision device are also stored in the RAM 1004. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; the storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the navigation route decision device to communicate wirelessly or wired with other devices to exchange data. Although the navigation route decision device with various systems is shown in the figure, it should be understood that all of the shown systems are not required to be implemented or possessed. More or less systems can be alternatively implemented or possessed.

[0082] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program according to embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network through the communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of embodiments of the present disclosure are performed.

[0083] The navigation route decision device provided by the present disclosure adopts the navigation route decision method in the above embodiments, and can solve the technical problem of the incomplete storage of historical travel data of the existing vehicle navigation system, which leads to the inability to effectively support memory-based navigation instructions. Compared with the prior art, the navigation route decision device provided by the present disclosure has the same beneficial effects as the navigation route decision method provided by the above embodiments, and other technical features in the navigation route decision device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0084] It should be understood that various parts of the present application can be implemented in hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0085] The above description is merely that of a specific implementation of the present application, and the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, and all such changes or replacements should be encompassed within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0086] The present application provides a computer readable storage medium having stored thereon computer readable program instructions (i.e., a computer program) for performing the navigation route decision method in the above embodiments.

[0087] The computer readable storage medium provided by the present application may, for example, be a U disk, but is not limited to an electric, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electric connection with one or more conductive wires, 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 above. In the present embodiment, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer readable storage medium can be transmitted by any appropriate medium, including but not limited to an electric wire, an optical cable, an RF (Radio Frequency), etc., or any suitable combination of the above.

[0088] The above computer readable storage medium can be contained in a navigation route decision device; or can exist separately without being assembled into a navigation route decision device.

[0089] The computer readable storage medium described above carries one or more programs, when the one or more programs are executed by the navigation route decision device, the navigation route decision device is caused to: in response to a navigation task instruction input by a user, extract key information in the navigation task instruction, the key information including at least one of semantic dimension key information, time dimension key information and geographic dimension key information; Based on the key information, the vehicle historical trip data in the preset vehicle end database that meets the key information is integrated into a candidate trip data set, wherein the vehicle historical trip data is obtained after structuring storage of at least one of trip time information, effective driving track, key passing point information and key passing point surrounding associated information; The navigation task instruction and the candidate trip data set are input into a preset large model for matching verification, and a corresponding navigation route is determined according to an output matching verification result.

[0090] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0091] The flow and block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flow and block diagrams can represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustrations, and combinations thereof, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or combinations of hardware and software.

[0092] The modules involved in the embodiments of the present application can be implemented in the form of software or in the form of hardware. In some cases, the name of the module does not constitute a limitation on the unit itself.

[0093] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions (i.e. computer program) for executing the above navigation route decision method, and can solve the technical problem of the incomplete history travel data storage of the existing vehicle navigation system, which leads to the inability to effectively support the memory-based navigation instruction. Compared with the prior art, the computer readable storage medium provided by the present application has the same beneficial effects as the navigation route decision method provided by the above-mentioned embodiments, which will not be repeated here.

[0094] The present application also provides a computer program product comprising a computer program, which, when executed by a processor, implements the steps of the navigation route decision method as described above.

[0095] The computer program product provided by the present application can solve the technical problem of the incomplete history travel data storage of the existing vehicle navigation system, which leads to the inability to effectively support the memory-based navigation instruction. Compared with the prior art, the computer program product provided by the present application has the same beneficial effects as the navigation route decision method provided by the above-mentioned embodiments, which will not be repeated here.

[0096] All the acquisition of signals, information or actions in the present application are carried out in accordance with the corresponding data protection regulations and policies of the country where the device is located, and with the authorization given by the corresponding device owner.

[0097] The above merely describes some embodiments of the present application, and does not limit the protection scope of the present application, and any equivalent structural transformation, direct / indirect application in other related technical fields made by using the content of the present application specification and drawings shall be included in the protection scope of the present application.

Claims

1. A navigation route decision-making method, characterized in that, The method includes: In response to a navigation task instruction input by a user, key information is extracted from the navigation task instruction, and the key information includes at least one of semantic key information, time key information and geographic key information; Based on the key information, the vehicle historical trip data that conforms to the key information in the preset vehicle database is integrated into a candidate trip dataset. The vehicle historical trip data is obtained by structuring and storing at least one of the following: trip time information, effective driving trajectory, key waypoint information, and key waypoint surrounding association information. The navigation task instructions and the candidate itinerary dataset are input into a preset large model for matching and verification. The corresponding navigation route is determined based on the output matching and verification results.

2. The method as described in claim 1, characterized in that, Before the step of integrating historical vehicle trip data matching the key information from the preset vehicle database into a candidate trip dataset, the following steps are included: Obtain vehicle map navigation routes, actual vehicle driving trajectory, and trip time information; Input the vehicle map navigation route and the actual driving trajectory of the vehicle into a preset large model to perform path fitting and deduplication of vehicle waypoint data to obtain the effective driving trajectory of the vehicle. Determine the key waypoint information in the effective driving trajectory of the vehicle and the corresponding extended surrounding association information of the key waypoints; The system stores at least one of the following information in a structured manner: the trip time information, the vehicle's effective driving trajectory, the key waypoint information, and the surrounding association information of the key waypoint as vehicle historical trip data and saves it to a preset vehicle-side database.

3. The method as described in claim 2, characterized in that, The key waypoint information includes key waypoint category information. Following the step of structurally storing at least one of the trip time information, the vehicle's effective driving trajectory, the key waypoint information, and the surrounding association information of the key waypoints as vehicle historical trip data and saving it to a preset vehicle-side database, the following steps are included: Semantic vectors are extracted from the category information and the surrounding association information of the key waypoints to obtain a semantic dimension vector index. Update the semantic dimension vector index to a preset local index library to match the semantic dimension key information.

4. The method as described in claim 2, characterized in that, Before the steps of obtaining vehicle map navigation routes, actual vehicle driving trajectories, and travel time information, the following are included: When a user is detected to have boarded the vehicle, the user's identity is verified, and the user's identity code is determined based on the verification result. Query whether the user's ID code authorizes the storage of vehicle history trip data; If authorized storage of vehicle history trip data has been established, a binding relationship is established between the currently stored vehicle history trip data and the user's identity code, and the user is notified via voice broadcast that trip data collection and encrypted storage will be performed.

5. The method according to any one of claims 1-4, characterized in that, The step of integrating historical vehicle trip data matching the key information from a preset vehicle database into a candidate trip dataset based on the key information includes: Obtain the current user's user ID code and confirm whether the user ID code authorizes the extraction of the current user's vehicle history trip data from the preset vehicle database; If authorized, based on the key information in the semantic dimension and combined with the preset local index library, the key waypoint information and the surrounding related information of the key waypoint in the vehicle's historical trip data are subjected to fuzzy matching in the semantic dimension to obtain the first candidate trip data that meets the semantic dimension. Based on key information in the time dimension, fuzzy matching is performed on the trip time information of the vehicle's historical trip data in the time dimension to obtain second candidate trip data that meets the time dimension requirements. Based on key information in the geographic dimension, the effective driving trajectory of the vehicle's historical trip data is subjected to fuzzy matching in the geographic dimension to obtain a third candidate trip data that conforms to the geographic dimension. The first candidate itinerary data, the second candidate itinerary data, and the third candidate itinerary data are aggregated into a temporary dataset; The trip data in the temporary dataset is deduplicated to obtain a deduplicated temporary dataset. Based on the key information from different dimensions, the association strength of the trip data in the deduplicated temporary dataset is assigned, and the trip data in the deduplicated temporary dataset is sorted according to the assignment results to obtain a candidate trip dataset sorted by association strength.

6. The method as described in claim 5, characterized in that, The trip data includes the vehicle's effective driving trajectory, the category information of key waypoints, and the surrounding association information of key waypoints; The step of assigning association strength scores to the trip data in the deduplicated temporary dataset and sorting the trip data in the candidate trip dataset according to the scoring results includes: The effective driving trajectory of the vehicle is assigned a first correlation strength score and the corresponding first weight is determined to obtain a first score; The category information of the key pathway points is assigned a second association strength score and the corresponding second weight is determined to obtain a second score. The third association strength score is assigned to the surrounding association information of the key route points and the corresponding second weight is determined to obtain the third score; Based on the first score, the second score, and the third score, a scoring result is determined, and the trip data in the candidate trip dataset is sorted according to the scoring result.

7. The method as described in claim 1, characterized in that, The step of determining the corresponding navigation route based on the output matching verification result includes: If the matching verification result indicates the existence of a key waypoint that matches the navigation task instruction, then the basic information of the key waypoint is retrieved, the key waypoint is set as the destination, and the corresponding navigation route is determined. If the matching verification result shows that there are two or more key waypoints that meet the navigation task instructions, then the list of candidate key waypoints is confirmed and the list of candidate key waypoints is output to the preset terminal so that the key waypoints selected by the preset terminal based on the list of candidate key waypoints are set as the destination and the corresponding navigation route is determined. If the matching verification result is that there is no key waypoint that matches the navigation task instruction, a notification will be made via voice broadcast; and / or, basic information of real-time surrounding related locations will be obtained, a real-time surrounding related location list will be constructed, and the candidate surrounding related location list will be output to a preset terminal, so that the surrounding related location selected by the preset terminal based on the candidate surrounding related location list will be set as the destination and the corresponding navigation route will be determined.

8. A navigation route decision-making device, characterized in that, The device includes: An extraction module is used to extract key information from a navigation task instruction input by a user in response to the instruction. The key information includes at least one of semantic key information, time key information, and geographic key information. The matching module is used to integrate the vehicle historical trip data that matches the key information in the preset vehicle database into a candidate trip dataset based on the key information. The vehicle historical trip data is obtained by structuring and storing at least one of the trip time information, effective driving trajectory, key waypoint information and key waypoint surrounding association information. The verification module is used to input the navigation task instructions and the candidate itinerary dataset into a preset large model for matching and verification, and to determine the corresponding navigation route based on the output matching and verification results.

9. A navigation route decision-making device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the navigation route decision method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the navigation route decision method as described in any one of claims 1 to 7.

11. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the navigation route decision method as described in any one of claims 1 to 7.

Images