Personalized semantic understanding system for assisting autonomous mobile platform

Abstract

The invention discloses a personalized semantic understanding system for assisting an autonomous mobile platform, and the system comprises a user personalized word meaning library which is used for recording the mapping relation between the user habit expression and the corresponding operation; the fuzzy semantic analysis module is used for identifying fuzzy expression in a natural language instruction of a user and generating a plurality of candidate targets or operation options of semantic generalization; the multi-round context modeling module is used for establishing a semantic context based on session history and performing disambiguation and complementation on current semantics; and the intention generation module is used for converting the processed semantic information into a structured instruction. The personalized semantic understanding system is reasonable in structural design, all the modules cooperate with one another, the effect that 1 + 1 is larger than 2 can be achieved, fuzzy semantic understanding and multi-round context modeling can be conducted, the personalized requirement of a user can be met, semantic understanding precision is high, user adaptability is good, and the personalized semantic understanding system is suitable for man-machine interaction of an autonomous mobile platform.

Description

Technical Field

[0001] This invention relates to the fields of artificial intelligence, human-computer interaction, natural language processing and assisted mobile devices, and to a personalized semantic understanding system for assisting autonomous mobile platforms. Background Technology

[0002] In intelligent control systems, language, as a natural interaction medium, has advantages such as strong intuitiveness and low user learning cost, and is therefore gradually being introduced into the human-computer interaction interface of various mobile platforms.

[0003] However, natural language itself is characterized by fuzzy expressions, significant individual differences, and strong context dependence, making accurate semantic understanding and parsing of user intent challenging. Semantic understanding systems based on rule-based or shallow learning methods often only handle clearly structured and standardized instructions, struggling to address instructions with ambiguous meanings, contextual dependence, or personalized vocabulary. In actual interactions, users tend to use vague and subjective language, and traditional semantic parsing systems lack context modeling and personalized learning capabilities, leading to inaccurate semantic understanding.

[0004] In recent years, large language models have made progress in open instruction understanding and multimodal semantic reasoning, providing a technological foundation for improving the natural language understanding capabilities of control systems. However, existing research mainly focuses on single-turn instruction execution, lacking systematic design for mechanisms such as "multi-turn semantic continuation," "user expression habit modeling," and "semantic offset monitoring," and has not yet formed a semantic understanding framework that continuously adapts to individual users.

[0005] Therefore, developing a personalized semantic understanding system with fuzzy semantic understanding, multi-turn contextual modeling capabilities, and high accuracy is of great practical significance. Summary of the Invention

[0006] Due to the aforementioned deficiencies in existing technologies, this invention provides a personalized semantic understanding system with fuzzy semantic understanding and multi-turn contextual modeling capabilities, as well as high understanding accuracy. This system possesses capabilities such as fuzzy semantic understanding, multi-turn contextual modeling, and user-personalized semantic database construction. When applied to an assisted autonomous mobile platform, it improves the accuracy of the platform in parsing natural language commands, overcoming the shortcomings of current semantic understanding systems that lack contextual modeling and personalized learning capabilities, and lack systematic design for mechanisms such as "multi-turn semantic continuation," "user expression habit modeling," and "semantic offset monitoring."

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A personalized semantic understanding system for assisting autonomous mobile platforms includes:

[0009] A user-personalized semantic database is used to record the mapping relationship between users' habitual expressions and their corresponding operations, so as to realize personalized semantic interpretation.

[0010] The fuzzy semantic parsing module is used to identify fuzzy expressions in user natural language commands and generate multiple candidate targets or operation options with semantic generalization.

[0011] The multi-turn contextual modeling module is used to build semantic context based on conversation history and to disambiguate and complete the current semantics.

[0012] The intent generation module is used to convert the processed semantic information into structured instructions and pass them to the control system of the assisted autonomous mobile platform.

[0013] Each module has a different function. The fuzzy semantic parsing module is responsible for identifying fuzzy expressions, semantic generalization, and generating candidate solutions (for example, generating multiple candidate target points based on visual input for "here" and "there," and proposing multiple possible options for "a little bit" and "a little bit"). The multi-turn contextual modeling module is responsible for recording multi-turn interaction history, extracting contextual features, and disambiguating and completing the current expression. The existence of a user-personalized word library allows users to set preference definitions for commonly used expressions, such as binding "home" to a specific coordinate. The system automatically learns and maintains user-defined vocabulary, expression preferences, nickname references, etc.

[0014] With the help of the above modules, the system can recognize ambiguous expressions in natural language commands, perform semantic completion and intent disambiguation by combining multi-turn interaction history and environmental context, and support semantic mapping of commonly used words by users to achieve personalized adaptation of expression style. The system outputs structured intent information, which can be used to assist mobile platforms in path planning or execution control operations.

[0015] The personalized semantic understanding system for assisting autonomous mobile platforms of the present invention has a reasonable structural design, and the modules work together to achieve a synergistic effect. It can not only perform fuzzy semantic understanding and multi-turn contextual modeling, but also meet the personalized needs of users. Its semantic understanding accuracy is high and its user adaptability is good. It is particularly suitable for human-computer interaction (language interaction) on autonomous mobile platforms and has good application prospects.

[0016] As a preferred technical solution:

[0017] As described above, a personalized semantic understanding system for assisting autonomous mobile platforms includes a user-personalized semantic database comprising:

[0018] User-defined units are used to support users in defining the mapping relationship between words and operations.

[0019] The automatic learning unit is used to establish a mapping relationship between natural language instructions, word meanings, and selection instructions based on the user's historical expression habits.

[0020] As described above, a personalized semantic understanding system for assisting autonomous mobile platforms involves inputting a user's natural language command into a personalized semantic database. The database then identifies the command. If a personalized semantic meaning exists within the command, the database replaces that part with the personalized semantic meaning. The processed data is then input into a fuzzy semantic parsing module. Otherwise, the command is directly input into the fuzzy semantic parsing module.

[0021] As described above, a personalized semantic understanding system for assisting autonomous mobile platforms includes a fuzzy semantic parsing module comprising:

[0022] The fuzzy expression recognition unit is used to process fuzzy words, including spatial references ("here", "over there", etc.), degree expressions ("a moment", "a little bit", etc.), and fuzzy time ("wait a while", "almost there", etc.).

[0023] The semantic generalization unit is used to perform semantic generalization on the processed fuzzy words and to score the generated semantic generalization candidate targets and operation options to guide subsequent reasoning.

[0024] As described above, a personalized semantic understanding system for assisting autonomous mobile platforms includes a contextual modeling module that automatically constructs a semantic graph based on conversation history to enhance the disambiguation capability of command intents, comprising:

[0025] The conversation history unit is used to store the semantic state and the semantics of frequently mentioned ambiguous words in consecutive interaction rounds;

[0026] The contextual reasoning unit is used to disambiguate and complete the current ambiguous expression by combining the conversation context, platform pose state, and task state.

[0027] As described above, a personalized semantic understanding system for assisting autonomous mobile platforms, wherein the intent generation module outputs structured results in the form of navigation targets, operation categories, confidence levels, etc., for controlling autonomous mobile platforms or smart devices.

[0028] The above technical solution is only one feasible technical solution of the present invention. The scope of protection of the present invention is not limited thereto. Those skilled in the art can reasonably adjust the specific design according to actual needs.

[0029] The above invention has the following advantages or beneficial effects:

[0030] (1) The personalized semantic understanding system for assisting autonomous mobile platforms of the present invention has a reasonable structural design and the modules work together to achieve a 1+1>2 effect;

[0031] (2) The personalized semantic understanding system for assisting autonomous mobile platforms of the present invention can not only perform fuzzy semantic understanding and multi-round contextual modeling, but also meet the personalized needs of users. Its semantic understanding accuracy is high and its user adaptability is good.

[0032] (3) The personalized semantic understanding system of the present invention for assisting autonomous mobile platforms is particularly suitable for human-computer interaction (language interaction) of various autonomous mobile platforms and has good application prospects. Attached Figure Description

[0033] The invention, its features, shape, and advantages will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings. Like reference numerals denote like parts throughout the drawings. The drawings are not drawn to scale; their focus is on illustrating the gist of the invention.

[0034] Figure 1 This is a system architecture block diagram of the personalized semantic understanding system for assisting autonomous mobile platforms according to the present invention;

[0035] Figure 2 This is a diagram illustrating the structure and working principle of the fuzzy expression parsing module;

[0036] Figure 3 A diagram illustrating the structure and working principle of the multi-turn context modeling module;

[0037] Figure 4 A diagram illustrating the structure and working principle of a personalized word definition database for users. Detailed Implementation

[0038] To make the objectives, technical solutions, beneficial effects, and significant advancements of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention.

[0039] Obviously, all the embodiments described are only some embodiments of the present invention, and not all embodiments; based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] Example 1

[0041] A personalized semantic understanding system for assisting autonomous mobile platforms is capable of processing fuzzy expressions, contextual dependencies, and personalized user expressions in natural language. It can convert user natural language commands into structured control intentions and transmit them to downstream navigation or control modules, such as... Figure 1 As shown, it includes:

[0042] A user-personalized semantic database is used to record the mapping relationship between users' habitual expressions and their corresponding operations, so as to realize personalized semantic interpretation.

[0043] User-personalized semantic database, such as Figure 4 As shown, it includes user-defined units for supporting users to autonomously define the mapping relationship between words and operations, and automatic learning units for establishing the mapping relationship between natural language instructions, word meanings and selection instructions based on the user's historical expression habits;

[0044] After a user inputs a natural language command into the user's personalized semantic database, the database identifies the command. If a personalized semantic meaning exists in the command, the database replaces the personalized semantic meaning in the command and then inputs the processed data into the fuzzy semantic parsing module. Otherwise, the command is directly input into the fuzzy semantic parsing module.

[0045] The fuzzy semantic parsing module is used to identify fuzzy expressions in user natural language commands and generate multiple candidate targets or operation options with semantic generalization.

[0046] Fuzzy semantic parsing module, such as Figure 2 As shown, it includes a fuzzy expression recognition unit for processing fuzzy words such as spatial reference, degree expression and fuzzy time, and a semantic generalization unit for semantic generalization of the processed fuzzy words and scoring the generated semantic generalization candidate targets and operation options.

[0047] The multi-turn contextual modeling module is used to build semantic context based on conversation history and to disambiguate and complete the current semantics.

[0048] Context modeling module, such as Figure 3 As shown, it includes a session history unit for storing the semantic state and the semantics of frequently mentioned ambiguous words in consecutive interaction rounds, and a contextual reasoning unit for disambiguating and completing the current ambiguous expression by combining the session context, platform pose state and task state.

[0049] The intent generation module is used to convert the processed semantic information into structured instructions (output in the form of navigation target, operation category, and confidence level) and pass them to the control system of the assisted autonomous mobile platform.

[0050] The operating logic of the personalized semantic understanding system used to assist autonomous mobile platforms is as follows:

[0051] Users input natural language commands via voice, text, or other means. The system first sends the input to the user's personalized word meaning database 103 for processing. When the user's input contains words that match entries in the word meaning database, the system will replace the words with the corresponding standard expressions. If no word meaning mapping is matched, the module is skipped, and the original input is preserved.

[0052] The personalized instructions are fed into the fuzzy semantic parsing module 101. Based on rules and semantic models, this module generalizes the fuzzy expression into a set of candidate targets or operation sequences and calculates a confidence score for each candidate to represent its semantic rationality.

[0053] Next, the candidate results are input into the multi-round contextual modeling module 102. This module consists of a conversation history unit and a contextual reasoning unit. It can construct a semantic context graph based on the current task status, platform location, time context and historical dialogue records, and perform contextual disambiguation and completion on the candidate options. For example, if the user mentioned "drying clothes" in the previous round, the current ambiguous expression "go there" will be more likely to be interpreted as "go to the balcony".

[0054] Finally, the processed semantic results enter the intent generation module 104, which integrates all parsing and disambiguation results into a structured control intent, including navigation target location, operation category, operation confidence, etc., for use by the mobile platform to assist in scheduling.

[0055] The modules of this system have both a logical order and conditional triggering and configurability, allowing adjustments based on different device configurations and user characteristics. Through this structure and linkage mechanism, the system can accurately understand users' natural language commands in real-world environments, making it particularly suitable for interactive scenarios with unclear expressions, habitual ambiguity, and high context dependence.

[0056] Verification has shown that the personalized semantic understanding system for assisting autonomous mobile platforms of the present invention has a reasonable structural design, with each module working together to achieve a synergistic effect greater than the sum of its parts. It can not only perform fuzzy semantic understanding and multi-turn contextual modeling, but also meet the personalized needs of users. Its semantic understanding accuracy is high and its user adaptability is good. It is particularly suitable for human-computer interaction (language interaction) on various autonomous mobile platforms and has good application prospects.

[0057] Those skilled in the art should understand that variations can be implemented by combining existing technology with the above embodiments, which will not be elaborated here. Such variations do not affect the essence of the present invention, and will not be elaborated here either.

[0058] The preferred embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and the devices and structures not described in detail should be understood as being implemented in a conventional manner in the art. Any person skilled in the art can make many possible variations and modifications to the technical solutions of the present invention using the methods and techniques disclosed above, or modify them into equivalent embodiments with equivalent changes, without departing from the scope of the present invention. This does not affect the essential content of the present invention. Therefore, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the present invention's technical solutions still fall within the protection scope of the present invention.

Claims

1. A personalized semantic understanding system for assisting autonomous mobile platforms, characterized in that: include: A user-personalized semantic database is used to record the mapping relationship between users' habitual expressions and their corresponding operations, so as to realize personalized semantic interpretation. The fuzzy semantic parsing module is used to identify fuzzy expressions in user natural language commands and generate multiple candidate targets or operation options with semantic generalization. The multi-turn contextual modeling module is used to build semantic context based on conversation history and to disambiguate and complete the current semantics. The intent generation module is used to convert the processed semantic information into structured instructions and pass them to the control system of the assisted autonomous mobile platform.

2. The personalized semantic understanding system for assisting autonomous mobile platforms according to claim 1, characterized in that, The user-personalized semantic database includes: User-defined units are used to support users in defining the mapping relationship between words and operations. The automatic learning unit is used to establish a mapping relationship between natural language instructions, word meanings, and selection instructions based on the user's historical expression habits.

3. A personalized semantic understanding system for assisting autonomous mobile platforms according to claim 2, characterized in that, After a user inputs a natural language command into the user's personalized semantic database, the database identifies the command. If a personalized semantic meaning exists in the command, the database replaces that part with the personalized semantic meaning. The processed data is then input into the fuzzy semantic parsing module. Otherwise, the command is directly input into the fuzzy semantic parsing module.

4. A personalized semantic understanding system for assisting autonomous mobile platforms according to claim 1, characterized in that, The fuzzy semantic parsing module includes: A fuzzy expression recognition unit is used to process fuzzy words, including spatial reference, degree expression, and fuzzy time; The semantic generalization unit is used to perform semantic generalization on the processed fuzzy words and to score the generated semantic generalization candidate targets and operation options.

5. A personalized semantic understanding system for assisting autonomous mobile platforms according to claim 1, characterized in that, The context modeling module includes: The conversation history unit is used to store the semantic state and the semantics of frequently mentioned ambiguous words in consecutive interaction rounds; The contextual reasoning unit is used to disambiguate and complete the current ambiguous expression by combining the conversation context, platform pose state, and task state.

6. A personalized semantic understanding system for assisting autonomous mobile platforms according to claim 1, characterized in that, The intent generation module outputs the structured results in the form of navigation target, operation category, and confidence level.

Images