Vehicle interaction method, apparatus, device, and storage medium

Abstract

This invention relates to the field of intelligent driving technology and discloses a vehicle interaction method, device, equipment, and storage medium. Responding to user interaction commands, the invention retrieves multiple candidate recommendation information from a preset vector database based on the user interaction commands; acquires current state information and, based on the current state information and user interaction commands, filters the multiple candidate recommendation information from a preset graph database to obtain multiple target recommendation information; and generates and executes vehicle interaction commands based on the multiple target recommendation information. This introduces the collaborative work of a preset vector database and a preset graph database. The preset vector database leverages its advantages in semantic understanding for initial retrieval, while the preset graph database utilizes its advantages in logical reasoning for filtering. This achieves the organic integration of vehicle state information with knowledge data in a preset knowledge base, improving the accuracy and logic of vehicle interaction.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of intelligent driving technology, specifically to a vehicle interaction method, device, equipment, and storage medium. Background Technology

[0002] With the application of large voice models in the automotive field, in-vehicle voice interaction systems have introduced data retrieval to achieve interaction with users. However, in in-vehicle systems, vehicle status data is usually stored in relational databases or real-time buses, while the knowledge data required by large models is usually stored in vector databases. Currently used data retrieval schemes cannot associate or merge the two types of data, resulting in limited accuracy and logic in vehicle interaction. Summary of the Invention

[0003] In view of the above problems, embodiments of the present invention provide a vehicle interaction method, apparatus, device and storage medium to solve the problem of limited accuracy and logic in vehicle interaction in the prior art.

[0004] According to one aspect of the present invention, a vehicle interaction method is provided, the method comprising: In response to the input of user interaction commands, the system searches a preset vector database based on the user interaction commands to obtain multiple candidate recommendation information. The preset vector database includes the context information corresponding to the user interaction commands and a preset knowledge base. Obtain current status information, and based on the current status information and user interaction instructions, filter multiple candidate recommendation information in the preset graph database to obtain multiple target recommendation information; Based on multiple target recommendation information, vehicle interaction commands are generated and executed.

[0005] In one optional implementation, based on user interaction commands, a preset vector database is searched to obtain multiple candidate recommendation information, including: Parse the user interaction commands to obtain the service requirement characteristics corresponding to the user interaction commands; Based on the context information corresponding to the user interaction command, the service requirement features are extended to obtain the extended requirement features corresponding to the service requirement features. Based on the extended demand features and the context information corresponding to user interaction commands, a search is performed in a preset knowledge base to obtain multiple candidate recommendation information.

[0006] In one optional implementation, based on current state information and user interaction commands, multiple candidate recommendation information are filtered in a preset graph database to obtain multiple target recommendation information, including: Determine the rules for the current scenario based on the current state information and user interaction instructions; Based on the current status information, the current vehicle node is determined in the preset map database; Based on the current scenario rules and the current vehicle node, multiple candidate recommendation information is filtered in the preset graph database to obtain multiple target recommendation information.

[0007] In one optional implementation, the current scenario rules are determined based on the current state information and user interaction instructions, including: Based on user preference information in the current state information, determine the first scenario rule; and / or, based on the control domain corresponding to the user interaction command, determine the second scenario rule; The current scenario rules are determined based on the preset scenario rules, the first scenario rules, and / or the second scenario rules.

[0008] In one optional implementation, vehicle interaction commands are generated and executed based on multiple target recommendation information, including: Based on multiple target recommendation information, the preset text template is filled to obtain the target interactive text; Based on the target interactive text, generate and execute vehicle interaction commands.

[0009] In one alternative implementation, the method further includes: Match user interaction commands with preset command templates to determine the command type of the user interaction command; If the user interaction command is of type 1, execute the vehicle interaction command corresponding to the user interaction command. If the user interaction command is of type 2, the step of retrieving multiple candidate recommendation information from the preset vector database based on the user interaction command is executed.

[0010] In one alternative implementation, the method further includes: In response to the current state information meeting the preset trigger conditions, the target interactive text is generated based on the current state information; Based on the target interactive text, generate and execute vehicle interaction commands.

[0011] According to another aspect of the present invention, a vehicle interaction device is provided, comprising: The recommendation information retrieval module is used to respond to the input of user interaction commands, and to retrieve multiple candidate recommendation information from a preset vector database based on the user interaction commands. The preset vector database includes the context information corresponding to the user interaction commands and a preset knowledge base. The recommendation information filtering module is used to obtain the current status information and, based on the current status information and user interaction instructions, filter multiple candidate recommendation information in a preset graph database to obtain multiple target recommendation information. The interaction command generation module is used to generate and execute vehicle interaction commands based on multiple target recommendation information.

[0012] According to another aspect of the present invention, an electronic device is provided, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus; The memory is used to store at least one executable instruction that causes the processor to perform operations of the vehicle interaction method as described in any of the above embodiments.

[0013] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein at least one executable instruction is stored therein, which, when executed on a vehicle interaction device or electronic device, causes the vehicle interaction device or electronic device to perform the operation of the vehicle interaction method as described in any of the preceding embodiments.

[0014] According to another aspect of the present invention, a computer program product is provided, including computer instructions for causing a vehicle interaction device or electronic device to perform operations of the vehicle interaction method as described in any of the above embodiments.

[0015] This invention introduces a pre-set vector database and a pre-set graph database working together. After a user interaction command is input, the pre-set vector database is searched based on the user interaction command to obtain multiple candidate recommendation information. Then, based on the current state information and the user interaction command, the multiple candidate recommendation information is filtered in the pre-set graph database to obtain multiple target recommendation information. This leverages the semantic understanding advantage of the pre-set vector database for preliminary retrieval, and the logical reasoning advantage of the pre-set graph database for further filtering. This achieves the organic integration of vehicle state information and knowledge data in the pre-set knowledge base, improving the accuracy and logic of vehicle interaction. At the same time, the context information corresponding to the user interaction command and the pre-set knowledge base are stored in the pre-set vector database, allowing the retrieval using the pre-set vector database to fully consider the historical interaction process between the user and the vehicle, further improving the accuracy and logic of vehicle interaction.

[0016] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0017] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a schematic flowchart of a first type of vehicle interaction method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a second process for a vehicle interaction method according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the third process of the vehicle interaction method according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the fourth process of the vehicle interaction method according to an embodiment of the present invention; Figure 5 This is a structural block diagram of a vehicle interaction device according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0018] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0019] With the application of large-scale speech models in the automotive field, in-vehicle voice interaction systems have incorporated data retrieval to enable interaction with users. However, in in-vehicle systems, vehicle status data is typically stored in relational databases or real-time buses, while the knowledge data required by large models is usually stored in vector databases. Currently used data retrieval solutions often only handle one type of data. For example, pure vector retrieval solutions excel at semantic understanding but cannot guarantee logical relationships between entities, while pure graph retrieval solutions excel at logical reasoning but struggle with fuzzy natural language queries. They cannot correlate or merge the two types of data, resulting in limited accuracy and logical consistency in vehicle interaction. Furthermore, current in-vehicle systems mostly passively respond to user interaction commands, lacking the ability to proactively push services and interactions to users.

[0020] To address the aforementioned technical problems, this invention provides a vehicle interaction method that introduces a preset vector database and a preset graph database to work collaboratively. After a user inputs an interaction command, the preset vector database is searched based on the command to obtain multiple candidate recommendations. Then, based on the current state information and the user's interaction command, the candidate recommendations are filtered in the preset graph database to obtain multiple target recommendations. This leverages the semantic understanding advantage of the preset vector database for initial retrieval, and the logical reasoning advantage of the preset graph database for further filtering. This achieves an organic integration of vehicle state information and knowledge data in a preset knowledge base, improving the accuracy and logic of vehicle interaction. Simultaneously, the preset vector database stores the context information corresponding to the user's interaction command and the preset knowledge base, ensuring that retrieval using the preset vector database fully considers the historical interaction process between the user and the vehicle, further enhancing the accuracy and logic of vehicle interaction.

[0021] According to an embodiment of the present invention, a vehicle interaction method embodiment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0022] This embodiment provides a vehicle interaction method. Figure 1 This is a schematic diagram of the first type of vehicle interaction method according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps: Step S101: In response to the input of the user interaction command, the preset vector database is searched based on the user interaction command to obtain multiple candidate recommendation information.

[0023] In this embodiment of the invention, the user interaction command can be a voice command input by the user via voice, or a non-voice command input by the user via touch, text input, or other means; no specific limitations are imposed here. When the user interaction command is a voice command, the vehicle terminal acquires the user's voice information in real time through its internal voice acquisition system, performs voice recognition processing on the acquired voice information, and converts it into text information.

[0024] In this embodiment of the invention, the preset vector database includes context information corresponding to user interaction commands and a preset knowledge base. The context information corresponding to user interaction commands includes historical dialogue content in the current session, i.e., user interaction commands input before the current round in multiple rounds of the current session, and vehicle interaction commands or recommendation information returned by the in-vehicle terminal. This information is used to accurately understand user semantics and dynamically inherit user preferences in the current session. The preset knowledge base is a vectorized storage of external knowledge data. It can be a knowledge base pre-loaded into the in-vehicle terminal or a knowledge base obtained and downloaded via the internet, used to provide knowledge support for semantic retrieval. The preset vector database stores relevant information in vector form to facilitate retrieval through vector similarity matching.

[0025] In this embodiment of the invention, in response to the input of a user interaction command, vector similarity matching is performed in a preset vector database based on the user interaction command, and a preset number of vector data that are semantically most similar to the user interaction command are obtained and used as candidate recommendation information.

[0026] In one optional implementation, after a user interaction command is input, the command type is first determined, and the subsequent processing flow is determined based on the command type, thereby improving interaction efficiency. Specifically, the user interaction command is matched with a preset command template to determine its command type. The preset command template includes command templates corresponding to multiple control domains, such as vehicle control domain templates, system control domain templates, vehicle information domain templates, and communication domain templates. The user interaction command is matched with the preset command templates. If the user interaction command matches any template in the preset command templates, the command type is determined to be a first type. If the user interaction command does not match any preset command template, the command type is determined to be a second type. The first type is a traditional control domain command type with a clear command intent. Based on the first type of user interaction command, a clear interactive feedback or control intent can be directly obtained. Therefore, if the command type of the user interaction command is the first type, the vehicle interaction command corresponding to the user interaction command is executed. The second type is the ambiguous instruction type with vague intent. Based on the second type of user interaction instruction, it is not possible to directly obtain a clear interactive feedback or control intent requirement. Therefore, if the instruction type of the user interaction instruction is the second type, the step of retrieving the preset vector database based on the user interaction instruction to obtain multiple candidate recommendation information is executed, thereby triggering the hybrid retrieval process of the present invention that combines the preset vector database and the preset graph database, so as to achieve accurate understanding of the user interaction instruction.

[0027] Step S102: Obtain the current status information, and based on the current status information and user interaction instructions, filter multiple candidate recommendation information in the preset graph database to obtain multiple target recommendation information.

[0028] In this embodiment of the invention, the current state information includes information representing the current environmental state, the current vehicle operating state, and the current occupant state. The current environmental state information may include information indicating the environmental state, such as in-vehicle temperature and humidity. The current vehicle operating state information may include information indicating the vehicle's operating status, such as the vehicle's current location, fuel / battery level, speed, and gear position. The current occupant state information may include information indicating the occupant's interaction status, such as their mood, actions, and posture. Optionally, the current state information may also include user profile information indicating user habits and preferences. The preset graph database stores state information and logical relationships between entities in the form of nodes and relational edges. Obtaining the current state information can be achieved by retrieving the state information stored in the node corresponding to the current vehicle at the current moment from the preset graph database.

[0029] In this embodiment of the invention, based on current state information and user interaction commands, multiple candidate recommendation information are logically verified and filtered in a preset graph database. Candidate recommendation information that does not conform to constraints such as physical scene, spatial constraints, or time validity indicated by the node relationships corresponding to the preset graph database is eliminated, resulting in multiple target recommendation information. For example, the preset graph database stores the location attributes of each node and its spatial distance relationship with vehicle nodes. By performing a graph query on the preset graph database, a set of nodes that are less than a preset distance threshold from the current location of the vehicle and are in operation can be filtered out. Candidate recommendation information whose corresponding nodes are located in the filtered node set is retained and used as target recommendation information.

[0030] Step S103: Based on multiple target recommendation information, generate and execute vehicle interaction instructions.

[0031] In this embodiment of the invention, based on multiple target recommendation information obtained after filtering, a vehicle interaction command is generated and executed. The vehicle interaction command can be of various types, such as voice broadcast commands, interface display commands, or control commands. Based on the multiple target recommendation information, a preset text template is filled to obtain target interaction text; based on the target interaction text, a vehicle interaction command is generated and executed.

[0032] In one optional implementation, the preset text template is a pre-configured response script template containing placeholders to be filled. Fields from multiple target recommendation messages are filled into the placeholders in the preset text template to generate complete target interaction text. After obtaining the target interaction text, corresponding voice broadcast commands or interface display commands are generated based on the target interaction text. These vehicle interaction commands are executed through the vehicle speakers or vehicle display screen to output interactive feedback to the user.

[0033] The vehicle interaction method provided in this invention introduces a preset vector database and a preset graph database to work together. After the user inputs an interaction command, the preset vector database is searched based on the user's interaction command to obtain multiple candidate recommendation information. Then, based on the current state information and the user's interaction command, the multiple candidate recommendation information is filtered in the preset graph database to obtain multiple target recommendation information. This leverages the semantic understanding advantage of the preset vector database for preliminary retrieval, and then utilizes the logical reasoning advantage of the preset graph database for further filtering. This achieves the organic integration of vehicle state information and knowledge data in the preset knowledge base, improving the accuracy and logic of vehicle interaction. At the same time, the preset vector database stores the context information corresponding to the user's interaction command and the preset knowledge base, enabling the retrieval using the preset vector database to fully consider the historical interaction process between the user and the vehicle, further improving the accuracy and logic of vehicle interaction.

[0034] This embodiment provides a vehicle interaction method. Figure 2 This is a schematic diagram of a second type of vehicle interaction method according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: Step S201: In response to the input of the user interaction command, the preset vector database is searched based on the user interaction command to obtain multiple candidate recommendation information.

[0035] Specifically, step S201 includes: Step S2011: Parse the user interaction command to obtain the service requirement characteristics corresponding to the user interaction command.

[0036] In this embodiment of the invention, service requirement features are used to characterize the type of service a user expects to obtain and the key parameters involved in that service. If the user interaction instruction is a text-based instruction input by the user, it can be directly parsed to extract the service requirement features. If the user interaction instruction is a voice-based instruction input by the user, it is first converted into text data, and then the text data is parsed to extract the service requirement features. The method for parsing the user interaction instruction can be semantic feature recognition and extraction, and no specific limitations are imposed here.

[0037] Step S2012: Based on the context information corresponding to the user interaction command, the service requirement features are extended to obtain the extended requirement features corresponding to the service requirement features.

[0038] In this embodiment of the invention, the context information corresponding to the user interaction command includes the historical dialogue content in the current session. Context information is obtained from a preset vector database; this context information consists of vector features of the historical dialogue content in the current session. The base context information may include interactive feedback information such as the destination confirmed by the user in previous interactions, expressed preferences, and rejected recommendations. Based on this interactive feedback information, the expansion of service demand features is constrained. Under this constraint, the service demand features are expanded by similar semantic meanings and supplemented with semantic content to obtain extended demand features.

[0039] Step S2013: Based on the extended demand features and the context information corresponding to the user interaction instructions, a search is performed in the preset knowledge base to obtain multiple candidate recommendation information.

[0040] In this embodiment of the invention, the extended demand features are transformed into vector features. Combined with the vector features of contextual information, an approximate nearest neighbor search is used to perform vector similarity matching in a preset knowledge base to recall a preset number of vector data that are semantically similar to the extended demand features as candidate recommendation information.

[0041] Step S202: Obtain the current state information, and based on the current state information and user interaction commands, filter multiple candidate recommendation information from the preset graph database to obtain multiple target recommendation information. For details, please refer to [link to relevant documentation]. Figure 1 Step S102 of the illustrated embodiment will not be described again here.

[0042] Step S203: Based on multiple target recommendation information, generate and execute vehicle interaction commands. For details, please refer to [link to relevant documentation]. Figure 1 Step S103 of the illustrated embodiment will not be described again here.

[0043] The vehicle interaction method provided in this invention parses user interaction commands to obtain service demand features and expands these features based on contextual information. This allows the retrieval process to fully utilize historical information from the current dialogue. Simultaneously, vector retrieval is performed in a preset knowledge base based on the expanded demand features and contextual information. The semantic understanding advantage of vector retrieval is used to obtain multiple semantically related candidate recommendation information, thereby ensuring the accuracy and reliability of the retrieval using the preset vector database.

[0044] This embodiment provides a vehicle interaction method. Figure 3 This is a schematic diagram of a third process of the vehicle interaction method according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps: Step S301: In response to the input of a user interaction command, a search is performed on a preset vector database based on the user interaction command to obtain multiple candidate recommendation information. For details, please refer to [link to relevant documentation]. Figure 1 Step S101 of the illustrated embodiment will not be described again here.

[0045] Step S302: Obtain the current status information, and based on the current status information and user interaction instructions, filter multiple candidate recommendation information in the preset graph database to obtain multiple target recommendation information.

[0046] Specifically, step S302 includes: Step S3021: Determine the current scenario rules based on the current status information and user interaction instructions.

[0047] In this embodiment of the invention, scene rules are used to characterize the constraints on which candidate recommendation information is filtered in the current scene, and may include constraints such as spatial distance constraints, state validity constraints, and temporal logic constraints. Based on the current state information and user interaction instructions, corresponding scene rules are determined respectively, and the determined scene rules are combined to obtain the current scene rules.

[0048] In one optional implementation, a first scenario rule is determined based on user preference information in the current state information; and / or, a second scenario rule is determined based on the control domain corresponding to the user interaction command. The user preference information includes user usage preferences and a summary and execution statistics of frequently used functions. This information can be obtained by summarizing and statistically analyzing historical user interaction feedback and vehicle control data. The preference content corresponding to the user preference information is assigned a higher recommendation weight to form the first scenario rule, which represents the user's personal preferences and selection tendencies. The control domain corresponding to the user interaction command can be determined through semantic analysis and intent recognition using a large language model. For user interaction commands that cannot be recognized by the large language model, their corresponding control domain is uncertain, and therefore the second scenario rule cannot be determined. In this case, the second scenario rule is not introduced. For user interaction commands that can be recognized by the large language model, after determining their corresponding control domain, the scenario rule corresponding to that control domain is determined as the second scenario rule. After determining the first scenario rule and / or the second scenario rule, the current scenario rule is determined based on the preset scenario rule and the first and / or second scenario rule. The preset scenario rule is a pre-configured, universally applicable standard rule. The preset scenario rule is then merged with the first and / or second scenario rule to obtain the current scenario rule. Optionally, when the preset scenario rule conflicts with the first or second scenario rule, the preset scenario rule takes precedence over the first and second scenario rules to ensure security and reliability.

[0049] Step S3022: Based on the current status information, determine the current vehicle node in the preset map database.

[0050] In this embodiment of the invention, based on the vehicle identifier or current location information in the current status information, the node corresponding to the current vehicle is located in a preset graph database and designated as the current vehicle node. The current vehicle node contains all the vehicle's real-time status attributes, providing a query starting point for subsequent graph traversal and filtering. Optionally, the current status information can be stored in the preset graph database in real time; that is, the current status information is used as the vehicle node and stored and updated in the preset graph database. During retrieval, the current vehicle node in the preset graph database is directly called without the need for a search.

[0051] Step S3023: Based on the current scene rules and the current vehicle node, filter multiple candidate recommendation information in the preset graph database to obtain multiple target recommendation information.

[0052] In this embodiment of the invention, multiple candidate recommendation messages retrieved from a preset vector database are input into a preset graph database. Using the current scene rules as filtering conditions and the current vehicle node as the query starting point, a graph traversal query is performed on the nodes corresponding to the multiple candidate recommendation messages in the preset graph database. Specifically, utilizing the logical relationships between nodes stored in the preset graph database, each candidate recommendation message is validated in multiple dimensions according to the current scene rules. For example, based on the spatial proximity relationship between nodes in the preset graph database and the current vehicle node, nodes corresponding to candidate recommendation messages whose distance from the current vehicle position is less than a distance threshold set in the current scene rules are filtered out. Similarly, based on the business status attributes of nodes in the preset graph database, nodes corresponding to candidate recommendation messages that meet the business status requirements in the current scene rules are filtered out, and so on. After the above graph traversal validation, candidate recommendation messages that meet the constraints corresponding to the current scene rules are determined as target recommendation messages.

[0053] Step S303: Based on multiple target recommendation information, generate and execute vehicle interaction commands. For details, please refer to [link to relevant documentation]. Figure 1 Step S103 of the illustrated embodiment will not be described again here.

[0054] The vehicle interaction method provided in this invention determines the current scene rules based on the current state information and user interaction instructions, determines the current vehicle node in a preset graph database based on the current state information, and performs graph traversal filtering on multiple candidate recommendation information based on the current scene rules and the current vehicle node. This fully utilizes the advantages of the preset graph database in logical reasoning, and the current scene rules enable the retrieval process using the preset graph database to simultaneously consider user preferences and control domain requirements, thereby eliminating semantically similar but logically unreasonable candidate information and improving the accuracy and reliability of the filtering results.

[0055] This embodiment provides a vehicle interaction method. Figure 4 This is a schematic diagram of the fourth process of the vehicle interaction method according to an embodiment of the present invention, such as... Figure 4 As shown, the process includes the following steps: Step S401: In response to the current state information satisfying the preset triggering conditions, generate target interactive text based on the current state information.

[0056] In this embodiment of the invention, the vehicle terminal continuously monitors the attributes of state nodes in a preset graph database to monitor in real time whether the current state information meets preset trigger conditions. The preset trigger conditions are pre-configured active service trigger rules, including trigger conditions related to time, location, and vehicle status. After obtaining the current state information, it is matched against the preset trigger conditions. If the current state information meets the preset trigger conditions, the active service process is triggered, and target interactive text is generated based on the current state information.

[0057] In one optional implementation, the preset triggering conditions may include vehicle status conditions and / or spatiotemporal conditions. Vehicle status conditions characterize the relevant conditions for triggering the active service, such as triggering the active service when fuel level is below 20%, battery level is below 30%, or tire pressure is abnormal. Spatiotemporal conditions characterize the relevant conditions for triggering the active service at the time or location, such as triggering the active service when the current time is during the weekday morning rush hour or the vehicle is located within a preset area. The preset triggering conditions may also be a combination of vehicle status conditions and spatiotemporal conditions, such as triggering the active service when "fuel level is below 20%, the current time is during the weekday morning rush hour, and the vehicle is located in a city center area."

[0058] In one optional implementation, the current state information is input into a large language model, which is then used to generate the target interactive text. Specifically, various state information items and user preference information from the current state information can be concatenated or combined to construct prompt words, which are then used to generate the target interactive text via the large language model.

[0059] Step S402: Generate and execute vehicle interaction instructions based on the target interactive text.

[0060] In this embodiment of the invention, vehicle interaction instructions are generated and executed based on the screened target interactive text. These vehicle interaction instructions can be of various types, such as voice broadcast instructions, interface display instructions, or control instructions.

[0061] The vehicle interaction method provided in this embodiment of the invention, in response to the current state information meeting the preset triggering conditions, generates target interaction text based on the current state information, and generates and executes vehicle interaction instructions based on the target interaction text. Thus, without the user actively inputting interaction instructions, it can proactively provide service suggestions to the user based on the real-time vehicle status, realize proactive service, and improve the intelligence level of vehicle interaction and user experience.

[0062] This embodiment also provides a vehicle interaction device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0063] This embodiment provides a vehicle interaction device, such as... Figure 5 As shown, it includes: The recommendation information retrieval module 501 is used to respond to the input of user interaction commands, and to retrieve multiple candidate recommendation information from a preset vector database based on the user interaction commands. The preset vector database includes the context information corresponding to the user interaction commands and a preset knowledge base. The recommendation information filtering module 502 is used to obtain the current status information and, based on the current status information and user interaction instructions, filter multiple candidate recommendation information in a preset graph database to obtain multiple target recommendation information. The interaction instruction generation module 503 is used to generate and execute vehicle interaction instructions based on multiple target recommendation information.

[0064] In one optional implementation, the recommendation information retrieval module 501 includes: The instruction parsing unit is used to parse user interaction instructions to obtain the service requirement characteristics corresponding to the user interaction instructions; The feature extension unit is used to extend the service requirement features based on the context information corresponding to the user interaction instructions, so as to obtain the extended requirement features corresponding to the service requirement features. The information retrieval unit is used to retrieve multiple candidate recommendation information from a preset knowledge base based on the extended demand features and context information corresponding to user interaction instructions.

[0065] In one optional implementation, the recommendation information filtering module 502 includes: The rule determination unit is used to determine the rules for the current scenario based on the current state information and user interaction instructions; The node determination unit is used to determine the current vehicle node in a preset map database based on the current state information; The information filtering unit allows users to filter multiple candidate recommendations from a preset graph database based on the current scene rules and the current vehicle node, thus obtaining multiple target recommendations.

[0066] In one optional implementation, the rule determination unit includes: The first rule determination subunit is used to determine the first scenario rule based on user preference information in the current state information; and / or, to determine the second scenario rule based on the control domain corresponding to the user interaction command; The second rule determination sub-unit is used to determine the current scenario rule based on the preset scenario rule, the first scenario rule, and / or the second scenario rule.

[0067] In one optional implementation, the interactive instruction generation module 503 includes: The text filling unit is used to fill a preset text template based on multiple target recommendation information to obtain the target interactive text; The instruction generation unit is used to generate and execute vehicle interaction instructions based on the target interactive text.

[0068] In one alternative embodiment, the device further includes: The type determination module is used to match user interaction commands with preset command templates to determine the command type of the user interaction command; The instruction execution module is used to execute the vehicle interaction instruction corresponding to the user interaction instruction if the instruction type of the user interaction instruction is the first type. The recommendation information retrieval module 501 is also used to perform a step of retrieving multiple candidate recommendation information from a preset vector database based on the user interaction command if the command type of the user interaction command is the second type.

[0069] In one alternative embodiment, the device further includes: The interactive text generation module is used to generate target interactive text based on the current state information in response to the current state information meeting the preset trigger conditions. The interactive instruction generation module 503 is also used to generate and execute vehicle interactive instructions based on the target interactive text.

[0070] The vehicle interaction device provided in this embodiment of the invention can execute the vehicle interaction method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the various modules and units described above are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0071] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0072] The following is a detailed reference. Figure 6 This diagram illustrates a suitable structural design for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 601, which can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) 602 or a program loaded from memory 608 into random access memory (RAM) 603. RAM 603 also stores various programs and data required for the operation of the electronic device. The processor 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0073] Typically, the following devices can be connected to I / O interface 605: input devices 606 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 607 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 608 including, for example, magnetic tapes, hard disks, etc.; and communication devices 609. Communication device 609 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 6 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0074] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory 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 via a communication device 609, or installed from a memory 608, or installed from a ROM 602. When the computer program is executed by the processor 601, it performs the functions defined in the vehicle interaction method of the embodiments of the present invention.

[0075] Figure 6 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention.

[0076] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the vehicle interaction method shown in the above embodiments is implemented.

[0077] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0078] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Furthermore, the embodiments of this invention are not directed to any particular programming language.

[0079] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. Similarly, for the sake of brevity and to aid in understanding one or more aspects of the invention, in the description of exemplary embodiments of the invention above, various features of the embodiments are sometimes grouped together in a single embodiment, figure, or description thereof. The claims, which follow the detailed description, are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0080] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components, except that at least some of such features and / or processes or units are mutually exclusive.

[0081] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A vehicle interaction method, characterized in that, The method includes: In response to the input of a user interaction command, a preset vector database is searched based on the user interaction command to obtain multiple candidate recommendation information, wherein the preset vector database includes context information corresponding to the user interaction command and a preset knowledge base; Obtain current status information, and based on the current status information and the user interaction command, filter multiple candidate recommendation information in a preset graph database to obtain multiple target recommendation information; Based on multiple target recommendation information, vehicle interaction commands are generated and executed.

2. The method according to claim 1, characterized in that, Based on the user interaction command, a search is performed on a preset vector database to obtain multiple candidate recommendation information, including: The user interaction command is parsed to obtain the service requirement characteristics corresponding to the user interaction command; Based on the context information corresponding to the user interaction command, the service requirement feature is extended to obtain the extended requirement feature corresponding to the service requirement feature. Based on the extended requirement features and the context information corresponding to the user interaction instructions, a search is performed in the preset knowledge base to obtain multiple candidate recommendation information.

3. The method according to claim 1, characterized in that, Based on the current state information and the user interaction instructions, multiple candidate recommendation information are filtered in a preset graph database to obtain multiple target recommendation information, including: Based on the current state information and the user interaction instructions, determine the current scenario rules; Based on the current status information, the current vehicle node is determined in the preset map database; Based on the current scene rules and the current vehicle node, multiple candidate recommendation information are filtered in the preset graph database to obtain multiple target recommendation information.

4. The method according to claim 3, characterized in that, The step of determining the current scenario rules based on the current state information and the user interaction instructions includes: Based on the user preference information in the current state information, a first scenario rule is determined; and / or, based on the control domain corresponding to the user interaction instruction, a second scenario rule is determined; The current scenario rule is determined based on the preset scenario rules, the first scenario rule, and / or the second scenario rule.

5. The method according to claim 1, characterized in that, The step of generating and executing vehicle interaction commands based on multiple target recommendation information includes: Based on multiple target recommendation information, a preset text template is filled to obtain the target interactive text; Based on the target interactive text, the vehicle interaction instructions are generated and executed.

6. The method according to claim 1, characterized in that, The method further includes: The user interaction command is matched with a preset command template to determine the command type of the user interaction command; If the instruction type of the user interaction instruction is the first type, execute the vehicle interaction instruction corresponding to the user interaction instruction; If the instruction type of the user interaction instruction is the second type, the step of retrieving the preset vector database based on the user interaction instruction to obtain multiple candidate recommendation information is executed.

7. The method according to claim 1, characterized in that, The method further includes: In response to the current state information satisfying a preset triggering condition, target interactive text is generated based on the current state information; Based on the target interactive text, the vehicle interaction instructions are generated and executed.

8. A vehicle interaction device, characterized in that, The device includes: The recommendation information retrieval module is used to respond to the input of user interaction instructions, and to retrieve multiple candidate recommendation information from a preset vector database based on the user interaction instructions. The preset vector database includes context information corresponding to the user interaction instructions and a preset knowledge base. The recommendation information filtering module is used to obtain current status information and, based on the current status information and the user interaction instructions, filter multiple candidate recommendation information in a preset graph database to obtain multiple target recommendation information. The interaction instruction generation module is used to generate and execute vehicle interaction instructions based on multiple target recommendation information.

9. An electronic device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction that causes the processor to perform the operation of the vehicle interaction method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The storage medium stores at least one executable instruction, which, when executed on a vehicle interaction device or electronic device, causes the vehicle interaction device or electronic device to perform the operation of the vehicle interaction method as described in any one of claims 1-7.

Images