Navigation interaction method and device, vehicle-machine interaction system and vehicle

Abstract

The application provides a navigation interaction method and device, a vehicle-machine interaction system and a vehicle. The method comprises the following steps: acquiring area information of each passing point in a current navigation path; determining the interest degree of a driver to each passing point based on the driver information and the area information of each passing point; the driver information is determined based on the historical driving data of the driver and / or the navigation operation data of the driver; and interacting with the driver based on the interest degree of the driver to each passing point. The navigation interaction method and device, electronic equipment and storage medium provided by the application can quickly and accurately determine the interest degree of the driver to each passing point through the driver information and the area information of each passing point, so as to realize personalized navigation interaction for the driver based on the interest degree of the driver to each passing point. The application not only can reduce the interaction communication cost, optimize the user experience, but also can improve the navigation efficiency and ensure the driving safety.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a navigation interaction method, device, vehicle-to-machine interaction system, and vehicle. Background Technology

[0002] As people's living standards improve, more and more people own private cars. When traveling long distances by private car, the routes are often complex, and it is necessary to get to the destination correctly. Therefore, in-car navigation systems have become an indispensable tool.

[0003] Currently, in-vehicle navigation systems are used to display map information and provide users with a variety of information functions, such as route guidance to their destination and traffic information near their current location. However, these functions are mostly general-purpose and cannot be adapted to users' preferences and needs. Therefore, they cannot fully meet user needs in route planning and guidance, i.e., they cannot provide personalized services. Summary of the Invention

[0004] This invention provides a navigation interaction method, device, vehicle-machine interaction system, and vehicle to address the shortcomings of existing in-vehicle navigation systems that cannot provide personalized services based on users' preferences and needs.

[0005] This invention provides a navigation interaction method, comprising:

[0006] Obtain the area information of each point along the current navigation path;

[0007] Based on driver information and the regional information of each waypoint, the driver's level of interest in each waypoint is determined; the driver information is determined based on the driver's historical driving data and / or the driver's navigation operation data.

[0008] The interaction is based on the driver's level of interest in each waypoint.

[0009] According to a navigation interaction method provided by the present invention, determining the driver's level of interest in each waypoint based on driver information and the area information of each waypoint includes:

[0010] Based on the current position in the current navigation path, determine the passed-through points and the points to be passed from each waypoint;

[0011] Based on the driver information, the stop information of the passed-through points, and the area information of the points to be passed, the driver's level of interest in the points to be passed is determined.

[0012] According to a navigation interaction method provided by the present invention, determining the driver's level of interest in the point to be passed based on the driver information, the dwell information of the passed-through points, and the area information of the point to be passed includes:

[0013] Based on the driver information, the stop information of the passed waypoints, the area information of the waypoints to be passed, and supplementary information, the driver's level of interest in the waypoints to be passed is determined; the supplementary information includes at least one of the driver's current driving time, the estimated time of arrival at the next waypoint, and the trip purpose, the trip purpose being determined based on the navigation origin and destination information and the driver information.

[0014] According to a navigation interaction method provided by the present invention, the step of interacting with the driver based on the driver's interest in various waypoints includes:

[0015] Determine the interaction time based on the current location or information broadcast by the navigation system;

[0016] At the interaction time, the interaction is conducted with the driver based on the driver's level of interest in each waypoint.

[0017] According to a navigation interaction method provided by the present invention, the regional information of each waypoint is determined based on the information broadcast by the navigation system.

[0018] According to a navigation interaction method provided by the present invention, the step of interacting with the driver based on the driver's interest in various waypoints includes:

[0019] If the driver's interest in any waypoint exceeds a preset threshold, the system mines the factors of the driver's interest in any waypoint based on the regional information of that waypoint and the driver's information, and interacts with the driver based on these factors of interest.

[0020] According to a navigation interaction method provided by the present invention, the step of interacting with the driver based on the driver's interest in various waypoints includes:

[0021] If the driver's interest in any waypoint exceeds a preset threshold, the navigation system broadcasts information based on that waypoint to interact with the driver.

[0022] According to a navigation interaction method provided by the present invention, the step of interacting with the driver based on the driver's interest in various waypoints includes:

[0023] Based on the navigation system's broadcast information at each waypoint and the driver's level of interest in each waypoint, the system interacts with the driver.

[0024] According to a navigation interaction method provided by the present invention, the step of interacting with the driver based on the driver's interest in various waypoints includes:

[0025] If the driver's interest in any waypoint is greater than a preset threshold, first introductory information for that waypoint is generated and played to the driver to interact with the driver.

[0026] If the driver's interest in any waypoint is less than or equal to the preset threshold, generate second introductory information for that waypoint and play the second introductory information to the driver to interact with the driver; or, if the driver's interest in any waypoint is less than or equal to the preset threshold, discard the interaction with the driver for that waypoint.

[0027] The text length of the first introductory information is greater than the text length of the second introductory information.

[0028] According to a navigation interaction method provided by the present invention, the driver information is determined based on the following steps:

[0029] Based on the driving route data in the historical driving data, the driver route information in the driver information is determined;

[0030] And / or,

[0031] Based on the driving trajectory data in the historical driving data and the navigation operation data, the driver identity information in the driver information is determined.

[0032] According to a navigation interaction method provided by the present invention, the current navigation path is determined based on the following steps:

[0033] Based on the navigation origin and destination information, navigation route planning is performed to generate multiple candidate navigation routes;

[0034] Based on the driver information, the current navigation path is determined from the multiple candidate navigation paths.

[0035] According to a navigation interaction method provided by the present invention, determining the current navigation path from a plurality of candidate navigation paths based on the driver information includes:

[0036] The driver information is matched with the navigation path information corresponding to each candidate navigation path to determine the matching degree of each candidate navigation path.

[0037] The candidate navigation path corresponding to the maximum matching degree is taken as the current navigation path.

[0038] The present invention also provides a navigation interaction device, comprising:

[0039] The acquisition unit is used to acquire the area information of each point along the current navigation path.

[0040] The determining unit is used to determine the driver's level of interest in each waypoint based on driver information and the area information of each waypoint; the driver information is determined based on the driver's historical driving data and / or the driver's navigation operation data;

[0041] An interaction unit is used to interact with the driver based on the driver's level of interest in each waypoint.

[0042] The present invention also provides a vehicle-machine interaction system, including: the navigation interaction device as described above.

[0043] The present invention also provides a vehicle, including: the vehicle-machine interaction system as described above.

[0044] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the navigation interaction method as described above.

[0045] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the navigation interaction method as described above.

[0046] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements any of the navigation interaction methods described above.

[0047] The navigation interaction method, device, vehicle-to-everything (V2X) system, and vehicle provided by this invention can quickly and accurately determine the driver's level of interest in each waypoint by using driver information and regional information of each waypoint. This enables personalized navigation interaction for the driver based on their level of interest in each waypoint, which not only reduces interaction and communication costs and optimizes user experience, but also improves navigation efficiency and ensures driving safety. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0049] Figure 1 This is a flowchart illustrating the navigation interaction method provided by the present invention;

[0050] Figure 2 This is a flowchart illustrating an implementation of step 120 in the navigation interaction method provided by the present invention.

[0051] Figure 3 This is a flowchart illustrating the current navigation path determination method provided by the present invention;

[0052] Figure 4 This is a schematic diagram of the navigation interaction device provided by the present invention;

[0053] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0055] Currently, in-vehicle navigation systems display map information and provide users with various information functions, such as route guidance to their destination and traffic information near their current location. However, these functions are mostly general-purpose and cannot be adapted to individual user preferences and needs. Therefore, their route planning and guidance cannot fully meet user requirements. Furthermore, in-vehicle navigation systems cannot provide personalized services, resulting in route planning and guidance that do not fully meet user needs. Users often need to view multiple candidate routes provided by the in-vehicle navigation system and choose the optimal route while driving, which is time-consuming, laborious, and poses driving safety hazards.

[0056] In response, the present invention provides a navigation interaction method. Figure 1 This is a flowchart illustrating the navigation interaction method provided by the present invention, as shown below. Figure 1 As shown, this method can be applied to in-vehicle navigation systems and includes the following steps:

[0057] Step 110: Obtain the area information of each point along the current navigation path.

[0058] Here, the current navigation route refers to the optimal route calculated by the in-vehicle navigation system based on the user-input start point, destination, and waypoints. Waypoints are points that need to be passed along the current navigation route, typically intermediate stops on the path from the start point to the destination. Each waypoint can include various locations and places, such as cities or administrative regions at different levels, streets, companies, offices, restaurants, hotels, tourist attractions, parks, etc. The regional information for each waypoint usually includes its geographical location, surrounding environment, and road traffic conditions. Geographical location refers to the longitude and latitude coordinates of the waypoint; the surrounding environment includes nearby buildings, shops, parks, hospitals, etc.; and road traffic conditions refer to the real-time traffic flow, congestion level, and construction status of the roads where the waypoints are located.

[0059] Optionally, the area information of each point along the current navigation path can be stored in the vehicle's storage device and then retrieved from the vehicle's storage device, or it can be stored on a cloud server and then retrieved from the cloud server.

[0060] Step 120: Based on driver information and regional information of each waypoint, determine the driver's level of interest in each waypoint; driver information is determined based on the driver's historical driving data and / or the driver's navigation operation data.

[0061] Specifically, driver information can be understood as personalized information about the driver, that is, information used to characterize the driver's interests, preferences, personality, etc. Driver information is determined based on the driver's historical driving data and / or navigation operation data. Historical driving data may include origin and destination, driving time, intermediate stops, intermediate stop duration, driving speed, road congestion, weather, etc., while navigation operation data may include data on the driver's actions in selecting the optimal route from multiple candidate routes, data on the driver's actions in switching routes during the drive, and data on the driver's actions in searching for specific scenarios.

[0062] For example, by analyzing intermediate stops and their durations in historical driving data, we can determine a driver's location preferences, such as whether they prefer shopping malls or parks. Similarly, by analyzing navigation operation data, which shows the selection of the optimal route from multiple candidate routes, we can determine a driver's route selection preferences, such as preferences for shorter travel times, shorter distances, fewer traffic lights, and less congestion.

[0063] Based on driver information and the regional information of each waypoint, the driver's level of interest in each waypoint can be determined. For example, based on the driver information, it can be known that the driver likes to go to parks. If the regional information of waypoint 'a' shows that there is a park near waypoint 'a', it can be judged that the driver has a high level of interest in waypoint 'a'.

[0064] Step 130: Interact with the driver based on their level of interest in each waypoint.

[0065] Specifically, after determining the driver's level of interest in each waypoint, the system can interact with the driver according to their level of interest. Optionally, if the driver's level of interest in a waypoint is greater than a preset threshold, detailed information about that waypoint can be played; if the driver's level of interest in a waypoint is less than or equal to the preset threshold, brief information about that waypoint can be played or no information about that waypoint can be played.

[0066] In some specific implementations, driver information and regional information of each waypoint can be input into a pre-built first interaction model. The first interaction model determines the driver's level of interest in each waypoint and generates waypoint information of varying lengths and levels of detail to broadcast to the driver based on the driver's level of interest, thereby achieving interaction with the driver. The first interaction model is trained based on sample driver information, regional information of sample waypoints, and interest tags of sample drivers for each sample waypoint and corresponding playback tags for each sample waypoint. The first interaction model can be built based on pre-trained models such as GPT (Generative Pre-trained Transformer), BERT (Bidirectional Encoder Representations from Transformers), XLNet (extremeMulti-label Learning Network), ROBERTa (Robustly Optimized BERT approach), and T5 (Text-to-Text Transfer Transformer).

[0067] Here, the first interaction model can also be a large-scale model deployed in a chatbot with human-like characteristics, such as ChatGPT (Chat Generative Pre-trained Transformer). A chatbot deployed with this first interaction model can engage in dialogue with users by understanding and learning human language, and can interact with users based on the context of the dialogue, possessing truly human-like communication abilities. In addition, it also possesses human capabilities such as editing, translation, and searching. Therefore, the first interaction model deployed within it must possess the ability to perform the aforementioned functions, such as speech-to-text transcription, semantic understanding, text matching, sentiment analysis, and intent recognition.

[0068] Therefore, the navigation interaction method provided in this embodiment of the invention can quickly and accurately determine the driver's level of interest in each waypoint by using driver information and regional information of each waypoint. Based on the driver's level of interest in each waypoint, personalized navigation interaction can be achieved for the driver, which can not only reduce the cost of interaction and communication and optimize the user experience, but also improve navigation efficiency and ensure driving safety.

[0069] Based on the above embodiments, Figure 2 This is a flowchart illustrating an implementation of step 120 in the navigation interaction method provided by the present invention, as shown below. Figure 2 As shown, step 120 includes:

[0070] Step 121: Based on the current position in the current navigation path, determine the passed-through points and the points to be passed from each waypoint;

[0071] Step 122: Based on driver information, stop information of passed points, and area information of pending point of interest, determine the driver's level of interest in the pending point of interest.

[0072] Specifically, the current navigation route typically includes multiple waypoints. As the driver follows the current navigation route, they will pass through these waypoints in order of proximity. After determining the current location, the driver can identify the waypoints already passed and those yet to be passed.

[0073] The stop information for already passed road points is used to characterize the driver's stopping behavior at these points, such as whether the driver stopped at the point and the duration of the stop. Based on the driver information, the stop information for already passed road points, and the area information for the road points to be passed, it is possible to analyze whether the driver is interested in the road points to be passed, that is, to determine the degree of the driver's interest in the road points to be passed.

[0074] For example, if the driver's information indicates that the driver likes to eat ramen, and the driver stayed at the previous stop (ramen shop) for 1 hour, if the next stop is also a ramen shop, since the driver has already stopped at the previous stop (ramen shop), it is highly unlikely that the driver will want to stop at the next stop (ramen shop). Therefore, it can be judged that the driver has a low level of interest in the next stop.

[0075] In some specific implementations, driver information, stop information of passed route points, and regional information of the route point to be passed can be input into a pre-built first interest recognition model. The first interest recognition model determines the driver's degree of interest in the route point to be passed. The first interest recognition model is trained based on sample driver information, stop information of passed sample route points, regional information of the sample sample route point to be passed, and the sample driver's interest labels for the sample route point to be passed. The first interest recognition model can be built based on pre-trained models such as GPT model, BERT model, XLNet model, ROBERTa model, and T5 model.

[0076] Based on any of the above embodiments, step 122 includes:

[0077] Based on driver information, stop information of passed waypoints, regional information of to-be-passed waypoints, and supplementary information, the driver's level of interest in the to-be-passed waypoints is determined. The supplementary information includes at least one of the driver's current driving time, estimated time of arrival at the next waypoint, and trip purpose, which is determined based on navigation origin and destination information and driver information.

[0078] Specifically, the supplementary information includes at least one of the following: the driver's current driving time, the estimated time of arrival at the next waypoint, and the purpose of the trip. The current driving time refers to the driving time at the current location. The estimated time of arrival at the next waypoint refers to the estimated time when the driver will arrive at the next waypoint, which can be estimated based on the distance to the next waypoint and the average driving speed. The purpose of the trip refers to the driver's purpose for the current navigation route, which can include leisure and vacation, adventure, business trip, etc.

[0079] In addition, the purpose of the trip can be determined based on the navigation origin and destination information and the driver's information. For example, if the navigation destination is a tourist attraction and the driver's information shows that the driver likes to travel, then it can be estimated that the purpose of the trip is a leisure vacation.

[0080] Based on driver information, stop information at previously passed points, regional information at upcoming points, and supplementary information, the driver's level of interest in the upcoming points can be determined. For example, if the driver information indicates they enjoy ramen, and they did not stop at the previous point (gas station), and the next point is also a gas station with a current driving time of 4 hours, given the long driving time and lack of rest at the previous point, it can be determined that the driver needs to rest, i.e., needs to stop at the next point (gas station). Therefore, it can be concluded that the driver has a high level of interest in the next point.

[0081] In some specific implementations, driver information, stop information of passed-through points, regional information and supplementary information of points to be passed can be input into a pre-built second interest recognition model. The second interest recognition model determines the driver's degree of interest in the points to be passed. The second interest recognition model is trained based on sample driver information, stop information of passed-through sample points, regional information and supplementary information of points to be passed sample points, and the sample driver's interest labels for the points to be passed sample points. The second interest recognition model can be built based on pre-trained models such as GPT, BERT, XLNet, ROBERTa, and T5.

[0082] Based on any of the above embodiments, step 130 includes:

[0083] Determine the interaction time based on the current location or information broadcast by the navigation system;

[0084] During the interaction, the system interacts with the driver based on their level of interest in each waypoint.

[0085] Typically, navigation systems announce information when the driver reaches a designated waypoint to facilitate interaction; the interaction time can be understood as the moment the driver arrives at the designated waypoint. Therefore, embodiments of the present invention can determine the distance between the current location and the designated waypoint, thereby determining the moment the driver arrives at the designated waypoint; or, based on the information announced by the navigation system, determine the current location, and then determine the interaction time accordingly. The announced information can be either voice or text.

[0086] At the moment of interaction, it indicates that the driver has reached the designated waypoint. At this time, the interaction is further conducted with the driver based on the driver's level of interest in the waypoint.

[0087] In some specific implementations, the current location or navigation system broadcast information can be input into the time determination model. The time determination model then determines the interaction time based on the current location or navigation system broadcast information, and at the interaction time, interacts with the driver based on the driver's interest in each waypoint. The time determination model can be built based on pre-trained models such as GPT, BERT, XLNet, ROBERTa, and T5.

[0088] Based on any of the above embodiments, determining the interaction time based on the navigation system broadcast information includes:

[0089] The navigation system broadcast information is input into the time determination model, which then determines the interaction time based on the semantic information of the corresponding broadcast text.

[0090] The timing determination model is trained based on sample broadcast information and the sample interaction times corresponding to the sample broadcast information.

[0091] Specifically, navigation systems typically announce information when the driver arrives at a waypoint. This announcement usually includes the place name of the waypoint. The time-determining model can then determine the location of the waypoint (the current location) based on this name. Furthermore, the model can determine the distance to the designated waypoint from the current location, thus determining the time when the driver arrives at the waypoint—the interaction time. This time-determining model can be built using pre-trained models such as GPT, BERT, XLNet, ROBERTa, and T5.

[0092] Furthermore, the broadcast information can be either voice or text. If the broadcast information is voice, the corresponding audio can be converted into broadcast text, and the current position can be determined based on the semantic information of the broadcast text, and the interaction time can be determined based on the current position. If the broadcast information is text, the corresponding broadcast text can be directly extracted, and the current position can be determined based on the semantic information of the broadcast text, and the semantic information of the interaction time can be determined based on the current position.

[0093] Based on any of the above embodiments, the regional information of each waypoint is determined based on the information broadcast by the navigation system.

[0094] Specifically, since navigation systems typically broadcast information when the driver arrives at a point along the way, the information broadcast usually includes the name of the point. This allows the driver to obtain the area information of the point based on its name.

[0095] In some specific implementations, navigation system broadcast information can be input into a human-vehicle interaction model. The human-vehicle interaction model determines the area information of the current waypoint based on the navigation system broadcast information and generates corresponding human-vehicle interaction data based on the area information of the current waypoint, so as to realize human-vehicle interaction with the driver. The human-vehicle interaction model is trained based on sample navigation system broadcast information and corresponding sample human-vehicle interaction data labels. The human-vehicle interaction model can be constructed based on pre-trained models such as GPT model, BERT model, XLNet model, ROBERTa model, and T5 model.

[0096] Based on any of the above embodiments, step 130 includes:

[0097] If a driver's interest in any waypoint exceeds a preset threshold, the system can mine the factors that the driver is interested in based on the regional information of any waypoint and the driver's information, and then interact with the driver based on these factors.

[0098] Specifically, if a driver's interest in any point along the route exceeds a preset threshold, it indicates a high level of interest. In this case, based on the location information and driver information of the point, factors of interest can be identified, and interaction with the driver can be initiated based on these factors. These factors of interest may include the local culture, current season, and local attractions of the point.

[0099] For example, if the driver is a travel enthusiast and the route point 'a' contains multiple attractions, then the driver's interest in the route point 'a' can be determined to be the local attractions. Therefore, detailed descriptions of the local attractions can be generated and broadcast to the current driver to achieve interaction with the driver.

[0100] In some specific implementations, when the driver's interest in any waypoint exceeds a preset threshold, the regional information of any waypoint and the driver's information can be input into a pre-built second interaction model. The second interaction model then uses the regional information and driver information of any waypoint to identify the factors of interest the driver has in any waypoint, and generates corresponding introductory information to be broadcast to the driver, thereby achieving interaction with the driver. The second interaction model is trained based on the regional information of sample waypoints, sample driver information, and the playback tags corresponding to the sample waypoints. The second interaction model can be built based on pre-trained models such as GPT, BERT, XLNet, ROBERTa, and T5.

[0101] Based on any of the above embodiments, step 130 includes:

[0102] If the driver's interest in any waypoint exceeds a preset threshold, the navigation system will broadcast information based on that waypoint and interact with the driver.

[0103] Specifically, if the driver's interest in any waypoint exceeds a preset threshold, it indicates a high level of interest in that waypoint. In this case, the navigation system's broadcast information for that waypoint can be used as interactive data to interact with the driver. The navigation system's broadcast information for any waypoint can be stored in the vehicle's storage device or on a cloud server; this embodiment of the invention does not impose specific limitations on this.

[0104] In some specific implementations, when the driver's interest in any waypoint exceeds a preset threshold, the navigation system's broadcast information for that waypoint is input into a first navigation interaction model. The first navigation interaction model then generates interactive data based on the navigation system broadcast information to enable interaction with the driver. This first navigation interaction model is trained based on sample navigation system broadcast information and sample interaction data. It can be constructed using pre-trained models such as GPT, BERT, XLNet, ROBERTa, and T5.

[0105] Based on any of the above embodiments, step 130 includes:

[0106] Based on the navigation system's broadcast information at each waypoint and the driver's level of interest in each waypoint, the system interacts with the driver.

[0107] Specifically, each waypoint may correspond to multiple navigation system broadcast messages of varying text lengths. For example, each waypoint may correspond to a detailed navigation system broadcast message with a longer text length, and a brief navigation system broadcast message with a shorter text length. If the driver has a high level of interest in any waypoint, the detailed navigation system broadcast message for that waypoint can be used as interactive data to interact with the driver; if the driver has a low level of interest in any waypoint, the brief navigation system broadcast message for that waypoint can be used as interactive data to interact with the driver.

[0108] In some specific implementations, the navigation system broadcast information for each waypoint and the driver's level of interest in each waypoint can be input into a second navigation interaction model. The second navigation interaction model then generates corresponding interaction data based on the navigation system broadcast information for each waypoint and the driver's level of interest, enabling interaction with the driver. Specifically, the second navigation interaction model is trained based on sample navigation system broadcast information for each waypoint, sample driver interest in each waypoint, and sample interaction data. The second navigation interaction model can be constructed based on pre-trained models such as GPT, BERT, XLNet, ROBERTa, and T5.

[0109] Therefore, the embodiments of the present invention can interact with the driver by broadcasting information from the corresponding navigation system based on the driver's interest in each waypoint, thereby achieving personalized navigation interaction for the driver.

[0110] Based on any of the above embodiments, step 130 includes:

[0111] If the driver's interest in any waypoint exceeds a preset threshold, first introductory information for that waypoint is generated and played to the driver to facilitate interaction.

[0112] If the driver's interest in any waypoint is less than or equal to a preset threshold, generate a second introduction to that waypoint and play it to the driver to interact with the driver; or, if the driver's interest in any waypoint is less than or equal to a preset threshold, discard the interaction with the driver for that waypoint.

[0113] The length of the first introductory information is greater than the length of the second introductory information.

[0114] Specifically, the text length of the first introductory information is greater than that of the second introductory information. This can be understood as the first introductory information being a detailed introduction to any given route point, and the second introductory information being a brief introduction to any given route point.

[0115] If the driver's interest in any waypoint exceeds a preset threshold, it indicates a high level of interest, and the first introductory information can be played. If the driver's interest in any waypoint is less than or equal to the preset threshold, it indicates a low level of interest, and the second introductory information can be played. Alternatively, interaction with the driver regarding that waypoint can be abandoned, meaning no introductory information for that waypoint can be played.

[0116] Based on any of the above embodiments, driver information is determined based on the following steps:

[0117] Based on the driving route data in historical driving data, determine the driver's route information in the driver information;

[0118] And / or,

[0119] Based on driving trajectory data and navigation operation data from historical driving data, the driver's identity information in the driver information is determined.

[0120] Specifically, driving route data can include origin and destination, driving time, intermediate stops, and other data. Driver route information is used to characterize the driver's personalized route information during the driving process, which can include route selection preferences (preferring short driving time, short driving distance, or fewer traffic lights and less congestion), location preferences (what kinds of places are they interested in, whether they prefer shopping malls or parks), and waypoint planning preferences (whether they have a habit of going to the restroom or buying snacks when driving long distances).

[0121] In some specific implementations, historical driving data can be input into a pre-built path information extraction model, which then extracts the driver's path information from the driving path data in the historical driving data. The path information extraction model is trained based on sample historical driving data and corresponding sample driver path information labels. The path information extraction model can be built based on pre-trained models such as GPT, BERT, XLNet, ROBERTa, and T5.

[0122] Driving trajectory data can include information such as the driver's speed, number of sharp turns, and number of sudden braking. Navigation operation data can include the driver's actions in selecting the optimal route from multiple candidate routes, switching routes during the drive, scene search operations, and actions performed by the driver on the in-vehicle infotainment system. For example, data on dozens of drivers' actions on the in-vehicle infotainment system could include their preferred music and content under various conditions such as highway driving, traffic congestion, sunny weather, and rainy weather, as well as their most frequent user requests. Driver identity information is used to represent the driver's personalized identity and may include the driver's personality.

[0123] In some specific implementations, historical driving data and navigation operation data can be input into a pre-built identity information extraction model, which then extracts the driver's identity information from the driving trajectory data and navigation operation data in the historical driving data. The identity information extraction model is trained based on sample historical driving data and corresponding sample driver identity information labels. The identity information extraction model can be built based on pre-trained models such as GPT, BERT, XLNet, ROBERTa, and T5.

[0124] Based on any of the above embodiments Figure 3 This is a flowchart illustrating the current navigation path determination method provided by the present invention, as shown below. Figure 3 As shown, the current navigation path determination method includes the following steps:

[0125] Step 310: Based on the navigation origin and destination information, perform navigation route planning and generate multiple candidate navigation routes;

[0126] Step 320: Based on the driver information, determine the current navigation path from multiple candidate navigation paths.

[0127] Specifically, after determining the navigation origin and destination information, navigation path planning can be performed based on the navigation origin and destination information to obtain multiple candidate navigation paths; alternatively, the navigation origin and destination information can be input into the navigation system, and the navigation system can generate multiple candidate navigation paths.

[0128] Each candidate navigation path is generated based on navigation origin and destination information. In order to meet the personalized needs of drivers, this embodiment of the invention selects the optimal candidate navigation path from each candidate navigation path based on driver information as the current navigation path, that is, the current path can meet the personalized needs of drivers.

[0129] In some specific implementations, navigation origin-end point information and driver information can be input into the path planning model. The path planning model generates multiple candidate navigation paths based on the navigation origin-end point information and determines the current navigation path from among the multiple candidate navigation paths based on the driver information. The path planning model is trained based on sample navigation origin-end point information, sample driver information, and sample navigation paths. The path planning model can be constructed based on pre-trained models such as GPT, BERT, XLNet, ROBERTa, and T5.

[0130] Based on any of the above embodiments, step 320 includes:

[0131] The driver's information is matched with the navigation path information corresponding to each candidate navigation path to determine the matching degree of each candidate navigation path.

[0132] The candidate navigation path with the highest matching degree is used as the current navigation path.

[0133] Specifically, navigation route information is used to characterize information such as waypoint planning, route scenarios, route time, route distance, number of traffic lights, and congestion level in the corresponding candidate navigation routes. Driver information can characterize the driver's preferences for navigation routes, such as preferences for waypoint planning, route scenarios, route time, route distance, number of traffic lights, and congestion level.

[0134] The driver's information is matched against the navigation path information corresponding to each candidate navigation path. The higher the matching degree, the better the corresponding candidate navigation path matches the driver. Therefore, using the candidate navigation path with the highest matching degree as the current navigation path enables personalized route planning for the driver, avoiding the driving safety hazards caused by the driver having to view multiple candidate paths provided by the in-vehicle navigation system and choose the optimal path while driving.

[0135] Based on any of the above embodiments, the present invention also provides a navigation interaction method, the method comprising:

[0136] First, historical driving data and navigation operation data are acquired. Then, driver route information is extracted from the driving route data in the historical driving data, and driver identity information is extracted from the driving trajectory data and navigation operation data in the historical driving data.

[0137] Next, based on the navigation origin and destination information, navigation route planning is performed, generating multiple candidate navigation routes, and the current navigation route is determined from the multiple candidate navigation routes based on the driver information.

[0138] After determining the current navigation route, based on the driver information and the regional information of each waypoint in the current navigation route, the system determines the driver's level of interest in each waypoint and interacts with the driver based on the driver's level of interest in each waypoint.

[0139] Furthermore, the system can determine the driver's level of interest in passing through certain points based on driver information, stop information of already passed points, regional information of points to be passed, and supplementary information. The supplementary information includes at least one of the driver's current driving time, estimated arrival time at the next point, and trip purpose, which is determined based on navigation origin and destination information and driver information.

[0140] It should be noted that the above steps can be implemented using a pre-built interaction model. For example, navigation origin and destination information, historical driving data, and navigation operation data can be directly input into the interaction model. The interaction model determines the driver information based on the historical driving data and navigation operation data, determines the current navigation route based on the navigation origin and destination information and the driver information, and determines the driver's level of interest in each waypoint based on the driver information and the regional information of each waypoint in the current navigation route. Then, it outputs interactive content for the driver based on the driver's level of interest in each waypoint. In other words, personalized navigation interaction for the current driver can be achieved through a single interaction model. Here, the interaction model can also be a large-scale model deployed in a chatbot with human-like characteristics, such as ChatGPT (Chat Generative Pre-trained Transformer). Chatbots deployed with this interaction model can understand and learn human language to converse with users, and can interact with users based on the context of the conversation, possessing truly human-like communication abilities. In addition, they also possess human abilities such as editing, translation, and searching. Therefore, the interactive models deployed within them must possess the capability to perform the aforementioned functions, such as speech transcription, semantic understanding, text matching, sentiment analysis, and intent recognition.

[0141] Each of the above steps can also be implemented using a corresponding pre-built model. For example, historical driving data and navigation operation data can be input into a pre-built information extraction model, which will extract driver information. Navigation origin and destination information and driver information can be input into a pre-built path planning model, which will determine the current navigation path. Driver information and regional information of each point along the current navigation path can be input into a pre-built navigation interaction model, which will determine the driver's level of interest in each point and interact with the driver based on that level of interest.

[0142] The navigation interaction device provided by the present invention is described below. The navigation interaction device described below and the navigation interaction method described above can be referred to in correspondence.

[0143] Based on any of the above embodiments, the present invention also provides a navigation interaction device, such as... Figure 4 As shown, the device includes:

[0144] The acquisition unit 410 is used to acquire the area information of each point along the current navigation path;

[0145] The determining unit 420 is used to determine the driver's level of interest in each waypoint based on the driver information and the area information of each waypoint; the driver information is determined based on the driver's historical driving data and / or the driver's navigation operation data;

[0146] The interaction unit 430 is used to interact with the driver based on the driver's level of interest in each waypoint.

[0147] Based on any of the above embodiments, the determining unit 420 includes:

[0148] The waypoint determination unit is used to determine the passed waypoints and the waypoints to be passed from the various waypoints based on the current position in the current navigation path;

[0149] An interest determination unit is used to determine the driver's level of interest in the destination based on the driver information, the stop information of the passed route points, and the area information of the destination to be passed.

[0150] Based on any of the above embodiments, the interest determination unit is used for:

[0151] Based on the driver information, the stop information of the passed waypoints, the area information of the waypoints to be passed, and supplementary information, the driver's level of interest in the waypoints to be passed is determined; the supplementary information includes at least one of the driver's current driving time, the estimated time of arrival at the next waypoint, and the trip purpose, the trip purpose being determined based on the navigation origin and destination information and the driver information.

[0152] Based on any of the above embodiments, the interaction unit 430 is used for:

[0153] The timing determination unit is used to determine the interaction time based on the current location or information broadcast by the navigation system;

[0154] The human-vehicle interaction unit is used to interact with the driver based on the driver's level of interest in each waypoint during the interaction time.

[0155] Based on any of the above embodiments, the regional information of each waypoint is determined based on the information broadcast by the navigation system.

[0156] Based on any of the above embodiments, the interaction unit 430 is used for:

[0157] If the driver's interest in any waypoint exceeds a preset threshold, the system mines the factors of the driver's interest in any waypoint based on the regional information of that waypoint and the driver's information, and interacts with the driver based on these factors of interest.

[0158] Based on any of the above embodiments, the interaction unit 430 is used for:

[0159] If the driver's interest in any waypoint exceeds a preset threshold, the navigation system broadcasts information based on that waypoint to interact with the driver.

[0160] Based on any of the above embodiments, the interaction unit 430 is used for:

[0161] Based on the navigation system's broadcast information at each waypoint and the driver's level of interest in each waypoint, the system interacts with the driver.

[0162] Based on any of the above embodiments, the interaction unit 430 is used for:

[0163] If the driver's interest in any waypoint is greater than a preset threshold, first introductory information for that waypoint is generated and played to the driver to interact with the driver.

[0164] If the driver's interest in any waypoint is less than or equal to the preset threshold, generate second introductory information for that waypoint and play the second introductory information to the driver to interact with the driver; or, if the driver's interest in any waypoint is less than or equal to the preset threshold, discard the interaction with the driver for that waypoint.

[0165] The text length of the first introductory information is greater than the text length of the second introductory information.

[0166] Based on any of the above embodiments, the device further includes:

[0167] The route information determination unit is used to determine the driver route information in the driver information based on the driving route data in the historical driving data;

[0168] And / or,

[0169] The identity information determination unit is used to determine the driver identity information in the driver information based on the driving trajectory data in the historical driving data and the navigation operation data.

[0170] Based on any of the above embodiments, the device further includes:

[0171] The candidate path determination unit is used to plan navigation paths based on navigation origin and destination information and generate multiple candidate navigation paths.

[0172] The current route determination unit is used to determine the current navigation route from the multiple candidate navigation routes based on the driver information.

[0173] Based on any of the above embodiments, the current path determination unit includes:

[0174] The matching degree determination unit is used to match the driver information with the navigation path information corresponding to each candidate navigation path to determine the matching degree of each candidate navigation path.

[0175] The navigation path determination unit is used to select the candidate navigation path corresponding to the maximum matching degree as the current navigation path.

[0176] Based on any of the above embodiments, the present invention also provides a vehicle-machine interaction system, including: a navigation interaction device as described in any of the above embodiments. The navigation interaction device can be wired to the main control unit of the vehicle-machine interaction system (e.g., using a wired interface such as USB or HDMI), or wirelessly connected to the main control unit of the vehicle-machine interaction system (e.g., via Bluetooth or WiFi).

[0177] Based on any of the above embodiments, the present invention also provides a vehicle, including: the vehicle-machine interaction system as described above.

[0178] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 5 As shown, the electronic device may include a processor 510, a memory 520, a communications interface 530, and a communications bus 540, wherein the processor 510, memory 520, and communications interface 530 communicate with each other via the communications bus 540. The processor 510 can call logical instructions in the memory 520 to execute a navigation interaction method, which includes: acquiring regional information of each waypoint in the current navigation path; determining the driver's level of interest in each waypoint based on driver information and the regional information of each waypoint; the driver information is determined based on the driver's historical driving data and / or the driver's navigation operation data; and interacting with the driver based on the driver's level of interest in each waypoint.

[0179] Furthermore, the logical instructions in the aforementioned memory 520 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0180] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, wherein when the program instructions are executed by a computer, the computer is able to execute the navigation interaction method provided by the above methods, the method comprising: acquiring regional information of each waypoint in the current navigation path; determining the driver's degree of interest in each waypoint based on driver information and the regional information of each waypoint; the driver information being determined based on the driver's historical driving data and / or the driver's navigation operation data; and interacting with the driver based on the driver's degree of interest in each waypoint.

[0181] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program is implemented to perform the navigation interaction methods provided above. The method includes: acquiring regional information of each waypoint in the current navigation path; determining the driver's degree of interest in each waypoint based on driver information and the regional information of each waypoint; the driver information being determined based on the driver's historical driving data and / or the driver's navigation operation data; and interacting with the driver based on the driver's degree of interest in each waypoint.

[0182] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0183] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0184] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A navigation interaction method, characterized in that, include: Obtain the area information of each point along the current navigation path; The waypoints refer to the points that need to be passed through in the current navigation path, and the waypoints include various locations and places; Based on driver information and the regional information of each waypoint, the driver's level of interest in each waypoint is determined. The driver information is determined based on the driver's historical driving data and / or the driver's navigation operation data; the historical driving data is used to determine the driver's location preferences, and the navigation operation data is used to determine the driver's route selection preferences. Based on the driver's level of interest in each waypoint, interact with the driver; The step of determining the driver's level of interest in each waypoint based on driver information and the area information of each waypoint includes: determining the waypoints already passed and the waypoints to be passed based on the current position in the current navigation path; Based on the driver information, the stop information of the passed-through points, the area information of the points to be passed, and supplementary information, the driver's level of interest in the points to be passed is determined.

2. The navigation interaction method according to claim 1, characterized in that, The determination of the driver's level of interest in each waypoint based on driver information and the regional information of each waypoint includes: Based on the current position in the current navigation path, determine the passed-through points and the points to be passed from each waypoint; Based on the driver information, the stop information of the passed-through points, and the area information of the points to be passed, the driver's level of interest in the points to be passed is determined.

3. The navigation interaction method according to claim 2, characterized in that, The step of determining the driver's level of interest in the proposed route points based on the driver information, the stop information of the passed route points, and the area information of the proposed route points includes: Based on the driver information, the stop information of the passed waypoints, the area information of the waypoints to be passed, and supplementary information, the driver's level of interest in the waypoints to be passed is determined; the supplementary information includes at least one of the driver's current driving time, the estimated time of arrival at the next waypoint, and the trip purpose, the trip purpose being determined based on the navigation origin and destination information and the driver information.

4. The navigation interaction method according to any one of claims 1 to 3, characterized in that, The interaction with the driver based on the driver's level of interest in each waypoint includes: Determine the interaction time based on the current location or information broadcast by the navigation system; At the interaction time, the interaction is conducted with the driver based on the driver's level of interest in each waypoint.

5. The navigation interaction method according to any one of claims 1 to 3, characterized in that, The regional information of each point along the route is determined based on the information broadcast by the navigation system.

6. The navigation interaction method according to any one of claims 1 to 3, characterized in that, The interaction with the driver based on the driver's level of interest in each waypoint includes: If the driver's interest in any waypoint exceeds a preset threshold, the system mines the factors of the driver's interest in any waypoint based on the regional information of that waypoint and the driver's information, and interacts with the driver based on these factors of interest.

7. The navigation interaction method according to any one of claims 1 to 3, characterized in that, The interaction with the driver based on the driver's level of interest in each waypoint includes: If the driver's interest in any waypoint exceeds a preset threshold, the navigation system broadcasts information based on that waypoint to interact with the driver.

8. The navigation interaction method according to any one of claims 1 to 3, characterized in that, The interaction with the driver based on the driver's level of interest in each waypoint includes: Based on the navigation system's broadcast information at each waypoint and the driver's level of interest in each waypoint, the system interacts with the driver.

9. The navigation interaction method according to any one of claims 1 to 3, characterized in that, The interaction with the driver based on the driver's level of interest in each waypoint includes: If the driver's interest in any waypoint is greater than a preset threshold, first introductory information for that waypoint is generated and played to the driver to interact with the driver. If the driver's interest in any waypoint is less than or equal to the preset threshold, generate second introductory information for that waypoint and play the second introductory information to the driver to interact with the driver; or, if the driver's interest in any waypoint is less than or equal to the preset threshold, discard the interaction with the driver for that waypoint. The text length of the first introductory information is greater than the text length of the second introductory information.

10. The navigation interaction method according to any one of claims 1 to 3, characterized in that, The driver information is determined based on the following steps: Based on the driving route data in the historical driving data, the driver route information in the driver information is determined; And / or, Based on the driving trajectory data in the historical driving data and the navigation operation data, the driver identity information in the driver information is determined.

11. The navigation interaction method according to any one of claims 1 to 3, characterized in that, The current navigation path is determined based on the following steps: Based on the navigation origin and destination information, navigation route planning is performed to generate multiple candidate navigation routes; Based on the driver information, the current navigation path is determined from the multiple candidate navigation paths.

12. The navigation interaction method according to claim 11, characterized in that, Determining the current navigation path from the multiple candidate navigation paths based on the driver information includes: The driver information is matched with the navigation path information corresponding to each candidate navigation path to determine the matching degree of each candidate navigation path. The candidate navigation path corresponding to the maximum matching degree is taken as the current navigation path.

13. A navigation interaction device, characterized in that, include: The acquisition unit is used to acquire the area information of each point along the current navigation path. The waypoints refer to the points that need to be passed through in the current navigation path, and the waypoints include various locations and places; The determining unit is used to determine the driver's level of interest in each waypoint based on the driver information and the area information of each waypoint; The driver information is determined based on the driver's historical driving data and / or the driver's navigation operation data; the historical driving data is used to determine the driver's location preferences, and the navigation operation data is used to determine the driver's route selection preferences. An interaction unit is used to interact with the driver based on the driver's level of interest in each waypoint; The step of determining the driver's level of interest in each waypoint based on driver information and the area information of each waypoint includes: determining the waypoints already passed and the waypoints to be passed based on the current position in the current navigation path; Based on the driver information, the stop information of the passed-through points, the area information of the points to be passed, and supplementary information, the driver's level of interest in the points to be passed is determined.

14. A vehicle-machine interaction system, characterized in that, include: The navigation interaction device as described in claim 13.

15. A vehicle, characterized in that, include: The vehicle-machine interaction system as described in claim 14.

16. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the navigation interaction method as described in any one of claims 1 to 12.

17. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the navigation interaction method as described in any one of claims 1 to 12.

Images