Dynamic process network construction method based on verb-noun joint modeling

Abstract

The invention provides a dynamic process network construction method based on verb-noun joint modeling, which analyzes complex behaviors through verb tetrads (main body, reference system, space-time dimension and property change), and combines noun dynamic association to construct a'verb-noun dynamic network '. Compared with traditional static node type association, the method achieves accurate description of the dynamic behavior evolution process. By means of a double-domain unified computing architecture, physical rules and subjective cognition are considered, and the system adaptability is improved. And in the aspect of quantization expression, innovations of representing an action eigenstate by a ground state vector, representing cross-culture action association by an entangled state, corresponding action conversion rules of quantum gate operation and the like are realized, and the modeling precision and efficiency are enhanced. Cross-domain knowledge migration is supported through a verb interface standardization form, for example, grabbing is migrated from industrial assembly to a medical operation scene. The method is suitable for the fields of automatic driving, financial transactions, medical diagnosis and the like, and provides theoretical and technical support for strong artificial intelligence development.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence and natural language processing technology, and in particular to a method for constructing a dynamic process network based on verb-noun joint modeling. The method parses and optimizes complex behaviors by clarifying the verb quadruple structure (subject, reference system, spatiotemporal dimensions, and property changes), and belongs to the field of basic artificial intelligence theory and application technology. Background Art

[0002] In existing technologies, artificial intelligence and natural language processing primarily rely on cognitive models centered around noun entities, such as node-based association methods like knowledge graphs. While these methods can represent relationships between static entities to a certain extent, they have significant shortcomings in describing dynamic behaviors and their evolution. This is particularly evident in the following points:

[0003] 1. Cognitive model limitations: Traditional AI approaches tend to view the world as a "collection of objects" rather than a "network of processes." This node-based approach makes it difficult to capture and express complex dynamic behaviors and the underlying process nature.

[0004] 2. Ambiguity in the spatiotemporal dimension: In many application scenarios, such as autonomous driving and industrial robot control, time and space are crucial for understanding and executing actions. However, existing technologies often fail to quantify and encode the spatiotemporal dimension as essential elements, resulting in inaccurate processing of temporal relationships.

[0005] 3. Difficulty in cross-domain transfer: Due to the lack of a unified action parsing framework, knowledge from different domains is difficult to effectively transfer. For example, the implementation logic of the action "grasping" in industrial assembly scenarios differs significantly from the requirements in medical surgery scenarios, but traditional methods cannot provide a universal and flexible modeling approach.

[0006] 4. Lack of explainability: Many current AI systems use black box models, making their decision-making processes difficult to understand and audit. This becomes a major obstacle in applications requiring high transparency, such as financial transactions and medical diagnosis.

[0007] To address these issues, this paper proposes a novel verb theory framework. The core of this theory lies in analyzing dynamic behavior through the four-tuple structure of verbs (subject, reference frame, spatiotemporal dimensions, and property changes), and constructing a complete "verb-noun dynamic network" based on the dynamic associations of nouns. This approach not only reveals the essence of intelligence—the ability to structuredly recognize dynamic relationships—but also provides a new theoretical foundation for the development of strong artificial intelligence.

[0008] Furthermore, verb theory achieves unified modeling of both the objective and subjective domains of reality through a dual-domain processing process (cognitive dependency domain (CDR) and non-cognitive domain (NCR). For example, in autonomous driving, the action of "avoidance" requires consideration of both the physical equations of motion (NCR) and the emotional coupling decision (CDR); in cross-cultural service robotics, the specific expressions of etiquette actions must be adjusted to suit different cultural contexts. This dual-domain unified computing architecture significantly improves the system's adaptability and robustness.

[0009] In summary, the shortcomings of existing technologies in dynamic behavior modeling provide broad research space and development prospects for this invention. By reconstructing the essential cognitive framework of actions, verb theory breaks through the cognitive paradigm and technical bottlenecks of traditional AI, and has important theoretical value and practical application significance. Summary of the Invention

[0010] 1. Theoretical Innovation

[0011] This paper proposes a method for constructing a dynamic process network based on verb-noun joint modeling, namely verb theory. The core of this theory is to analyze and optimize complex behaviors through the four-tuple structure of verbs (subject, reference system, spatiotemporal dimension, and property change), and to construct a complete "verb-noun dynamic network" by combining the dynamic association of nouns. Specifically:

[0012] 1. Verb definition: A verb is defined as a four-tuple structure (subject, reference system, time and space dimension, property change), in which each element is a noun or a nominal entity.

[0013] 2. Verb-noun joint modeling:

[0014] Verb theory not only focuses on the verb itself, but also emphasizes the dynamic relationship between verbs and nouns. Specifically,

[0015] Verbs depend on nouns: The meaning and function of verbs need to be concretized through nouns (such as subject, reference system, and property change).

[0016] Nouns rely on verbs: nouns form a dynamic relationship network under the influence of verbs, giving static entities dynamic meaning.

[0017] Joint modeling: Through the interaction of the verb quadruple structure and related nouns, a complete "verb-noun dynamic network" is constructed.

[0018] 2. Technical Implementation

[0019] (1) Dual-domain unified computing architecture (such as Figure 2 )

[0020] 1. Physical domain verification

[0021] In the present invention, physical domain verification is performed by quantum Monte Carlo simulation to check the physical feasibility of the action. Specifically:

[0022] Theoretical foundation: Based on the objective reality in the non-cognitive domain (NCR), including basic physical layers, formal systems, and pre-observational entities. For example, the objective laws of space-time structure, fundamental interactions, and material motion.

[0023] Technical Implementation: Quantum Monte Carlo methods are used to verify the physical constraints of actions in real time. For example, in autonomous driving scenarios, the verb "avoid" requires verifying that the curvature of the vehicle's trajectory conforms to the physical equations of motion. In the field of industrial robotics, the verb "grasp" requires ensuring that the contact force between the robotic arm and the part meets the conditions of mechanical equilibrium.

[0024] Practical Application: Tensor cores are used to verify the physical feasibility of actions, ensuring that all dynamic behaviors conform to the laws of nature. For example, in surgical action risk prediction, the verb "resection" requires verification of whether the instrument force exceeds the specified limit.

[0025] 2. Cognitive Domain Reasoning

[0026] Cognitive domain reasoning uses hypergraph neural networks to analyze the meaning of actions in cultural contexts and enhance adaptability to subjective cognitive domains. Specific details are as follows:

[0027] Theoretical basis: Based on the knowledge system in the cognitive dependency domain (CDR), it includes the individual level (S), the group level (OS), and the social level (IS). These levels correspond to personal perceptual experience, collective cognitive paradigm, and macro-cognitive structure, respectively.

[0028] Technical Implementation: Hypergraph neural networks are used to capture the cultural context of actions. For example, in the field of cross-cultural service robots, the action of "handing something over" is represented by both hands and a 15-degree bow in the Japanese model, while it is represented by the right hand and avoiding eye contact in the Middle Eastern model.

[0029] Practical application: Dynamically adjusting the specific forms of actions in different cultural contexts. For example, in the case of etiquette adaptation, quantum entanglement can be used to quickly switch cultural rules, ensuring that service robots can provide appropriate etiquette movements based on the cultural habits of different regions.

[0030] 3. Innovation in Quantum Expression

[0031] This invention achieves multiple technical innovations in quantization expression, including the following three key points:

[0032] (1) The base state vector represents the action eigenstate

[0033] Theoretical basis: The four-tuple structure of verbs is mapped into quantum Hilbert space, with the ground state vector representing the eigenstate of the action. For example, the verb "cut" can be represented as |cut>, where the state includes the subject (the knife), the reference frame (the wood), the spacetime dimension (linear time), and the property change (the change in molecular structure).

[0034] Technical implementation: The state space of actions is constructed through the principle of quantum state superposition. For example, in the optimization of flexible assembly actions, the state of the verb "grasp" can be represented by quantum state superposition as a linear combination of |grasp successfully> and |slip>:

[0035] |ψ>=α|successful grasp>+β|slip>

[0036] Practical application: In the high-frequency trading verb map, a Hilbert space of trading verbs is constructed, and verb combinations are optimized through the Grover algorithm, significantly improving the speed of identifying arbitrage opportunities.

[0037] (2) Entangled states represent cross-cultural action associations

[0038] Theoretical basis: Quantum entanglement can be used to describe the associations between actions in different cultural contexts. For example, the differences between Chinese and Western bowing can be expressed through entanglement, reflecting the different understandings of the same action in two cultures.

[0039] Technical Implementation: Dynamic switching of cultural rules is achieved through quantum entanglement. For example, in the case of a cross-cultural service robot, the hardware supports the CDR inference engine to run cultural rules with ultra-low power consumption, while the NCR verification core ensures that physical movements do not violate ergonomics.

[0040] Practical application: In etiquette action adaptation, the robot can automatically adjust the action expression according to the cultural background of the user's region, thereby improving the user experience.

[0041] (3) Quantum gate operation corresponding action conversion rules

[0042] Theoretical basis: Mapping the action conversion rules into quantum gate operations. For example, the energy level transition from "walking to running" can be achieved through a specific quantum gate.

[0043] Technical Implementation: Quantum gate operations are used to define the transition logic between actions. For example, in an autonomous driving emergency obstacle avoidance decision-making system, the trajectory change of the "avoid" verb can achieve optimal path selection through a series of quantum gate operations.

[0044] Practical application: In the field of medical diagnosis, in the risk entropy calculation of the verb "resection", quantum gate operations are used to dynamically adjust the threshold range of the instrument's force, thereby reducing the risk of surgical actions.

[0045] (2) Specific modeling methods

[0046] 1. Definition of verb tetragram structure (e.g. Figure 1 )

[0047] Verbs are defined in this invention as the following four-tuple:

[0048] Verb = [Subject, Reference, Time / Space, Property] (1) Subject (moving subject)

[0049] Theoretical basis: The subject of motion is the entity or object that performs the action, usually represented by a noun. For example, in the verb "cut", the knife is the subject.

[0050] Technical implementation: The meaning and function of a verb are specified by specifying its subject. For example, in the field of industrial robotics, the subject of the verb "grab" is the robotic arm; in the field of financial trading, the subject of the verb "go long" is the investor.

[0051] Practical application: The choice of subject directly affects the feasibility and effectiveness of the action. For example, in an autonomous driving emergency obstacle avoidance decision-making system, the subject of the "avoid" verb is the vehicle.

[0052] (2)Reference (reference system)

[0053] Theoretical basis: A reference frame is the object or environment in which an action is performed. It provides the context in which the action occurs and is usually a noun. For example, in the verb "cut," wood serves as the reference frame.

[0054] Technical Implementation: The reference frame defines the specific scope and goal of an action. For example, in medical diagnosis, the reference frame for the verb "remove" is the tumor boundary; in industrial assembly scenarios, the reference frame for the verb "calibrate" is the datum surface.

[0055] Practical application: The choice of reference frame determines the accuracy of the action. For example, in surgical action risk prediction, the reference frame of the verb "suture" is the elastic modulus of biological tissue.

[0056] (3) Time / Space

[0057] Theoretical basis: The spatiotemporal dimension describes the temporal and spatial scope of an action and can be considered an abstract nominal entity. For example, in the verb "cut," linear time serves as the spatiotemporal dimension.

[0058] Technical Implementation: By quantifying the spatiotemporal dimensions, we ensure that actions conform to physical laws and cultural contexts. For example, in autonomous driving scenarios, the spatiotemporal dimension of the verb "avoid" is the change in trajectory curvature within 0.5 seconds; in the high-frequency trading verb map, the spatiotemporal dimension of the verb "go long" is the price volatility within a specific time window.

[0059] Practical Application: Accurate modeling of the spatiotemporal dimensions is crucial for optimizing dynamic behavior. For example, in the optimization of flexible assembly motions, the spatiotemporal dimensions of the “grasp” verb are the calculation of contact forces in 3D space.

[0060] (4) Property

[0061] Theoretical basis: A property change refers to a change in state or property caused by an action and can be expressed as a noun. For example, in the verb "cut," the change in molecular structure is considered a property change.

[0062] Technical Implementation: By defining property changes, the specific effects and goals of actions are clearly defined. For example, in the field of medical surgery, the property change of the verb "remove" becomes organ removal; in the field of industrial robotics, the property change of the verb "calibrate" becomes pixel matching.

[0063] Practical application: Accurate description of property changes helps evaluate the effectiveness of actions. For example, in surgical action risk prediction, the property change of the verb "suture" is tension control error.

[0064] 2. Verb-noun joint modeling

[0065] Verb theory not only focuses on the verbs themselves, but also emphasizes the dynamic relationship between verbs and nouns. The following is a specific modeling method:

[0066] (1) Verbs depend on nouns

[0067] Theoretical basis: The meaning and function of verbs need to be concretized through nouns (such as subject, reference frame, spatiotemporal dimension, and property change). For example, the verb "cut" depends on the knife (subject), wood (reference frame), linear time (spatiotemporal dimension), and molecular structure change (property change).

[0068] Technical Implementation: By clarifying the four-tuple structure of verbs, abstract verbs are transformed into specific action descriptions. For example, in an autonomous driving scenario, the verb "avoid" depends on the vehicle (subject), the obstacle (reference frame), 0.5 seconds (space-time dimension), and the change in trajectory curvature (property change).

[0069] Practical application: The dependency of verbs on nouns ensures the specificity and operability of actions. For example, in the field of cross-cultural service robotics, the verb "hand something" relies on using both hands / a 15-degree bow (Japanese model) or right hand / avoiding eye contact (Middle Eastern model).

[0070] (2) Nouns rely on verbs

[0071] Theoretical basis: Nouns form dynamic relational networks under the influence of verbs, giving dynamic meaning to static entities. For example, the noun "knife" acquires dynamic meaning through the verb "cut."

[0072] Technical Implementation: Through the action of verbs, static nouns are transformed into dynamic relational networks. For example, in industrial assembly scenarios, the "robotic arm" forms a dynamic relationship with the part through the verb "grab."

[0073] Practical Application: The noun-verb-based approach enhances the adaptability and flexibility of the system. For example, in the field of medical diagnosis, "surgical instruments" dynamically interact with tumor boundaries through the verb "resection."

[0074] (3) Joint modeling

[0075] Theoretical basis: Through the interaction of the four-tuple structure of verbs and related nouns, a complete "verb-noun dynamic network" is constructed. For example, the verb "cut" combines knife, wood, linear time and molecular structure change.

[0076] Technical Implementation: By establishing a verb-noun association matrix, comprehensive analysis and optimization of dynamic behaviors are achieved. For example, in autonomous driving scenarios, the "avoid" verb can be used to calculate trajectory changes in real time through the association matrix.

[0077] Practical Applications: Joint modeling significantly improves system performance and efficiency. For example, in automotive welding scenarios, verb theory significantly reduces the error rate of actions; in high-frequency trading verb graphs, optimizing verb combinations through the Grover algorithm significantly increases the speed of identifying arbitrage opportunities.

[0078] 3. Verb Interface for Automatic Knowledge Transfer

[0079] (1) Definition and function of verb interface

[0080] The Verb Interface, based on the core concepts of verb theory, establishes a standardized framework for knowledge representation and transfer by clarifying the four-tuple structure of verbs (subject, reference frame, spatiotemporal dimensions, and property changes) and the dynamic associations of nouns. Its primary function is to provide a unified modeling approach, enabling knowledge from different domains to be expressed in a consistent manner and enabling cross-domain transfer by adjusting specific parameters.

[0081] The standardized form and core of the verb interface lies in the four-tuple structure of the verb:

[0082] Verb=[Subject, Reference, Time / Space, Property]

[0083] Subject: The entity or object that performs an action.

[0084] Reference: The object or environment in which an action is performed.

[0085] Time / Space: Describes the time and space range in which an action occurs.

[0086] Property: A change in state or property caused by an action.

[0087] Through this standardized structure, the verb interface is able to decompose complex dynamic behaviors into a set of clear elements, thus providing a basis for cross-domain knowledge transfer.

[0088] (2) Specific mechanism of automatic knowledge transfer using verb interface

[0089] 1. Unified modeling capabilities

[0090] The verb interface achieves unified modeling of dynamic behaviors by clarifying the four-tuple structure of verbs. Regardless of the field, as long as the action can be decomposed into four elements: subject, reference frame, spatiotemporal dimension, and property change, it can be expressed through the verb interface. For example:

[0091] Industrial Robotics Field:

[0092] “Grasping” = [Robot arm (main body) → Part (reference frame) → 3D space (space-time dimension) → Contact force (property change)]

[0093] Medical surgery field:

[0094] “Grasping” = [surgical instrument (subject) → tissue (reference frame) → time accuracy (space-time dimension) → tension control error (property change)]

[0095] Although the specific parameters of the two scenarios are different, their core structures remain consistent, which provides the possibility of cross-domain migration.

[0096] 2. Parameterized Adjustment

[0097] The verb interface achieves adaptation to different domains through parameterized adjustments. Specifically, each element in the verb's quadruple structure can be flexibly adjusted according to the characteristics of the target domain. For example:

[0098] In the field of autonomous driving, the four-tuple structure of the "avoid" verb is:

[0099] [Vehicle (subject) → obstacle (reference frame) → 0.5 seconds (space-time dimension) → trajectory curvature changes (property changes)]

[0100] In financial services, the verb "avoid" can be adjusted to:

[0101] [Investor (subject) → Risk asset (reference frame) → Time window (time and space dimension) → Price volatility (nature change)]

[0102] In this way, the verb interface can quickly adapt to different application scenarios without changing the core structure.

[0103] 3. Dual-domain unified computing architecture

[0104] The Verb Interface not only considers the physical laws of the Non-Cognitive Domain (NCR), but also takes into account the subjective cognitive factors of the Cognitive Dependent Domain (CDR). This dual-domain unified computing architecture enables the Verb Interface to flexibly switch between the objective reality domain and the subjective cognitive domain. For example:

[0105] In the field of autonomous driving, “avoidance” actions require verification of whether the change in trajectory curvature conforms to the physical equation of motion (NCR), while also considering emotional coupling decision making (CDR).

[0106] In the field of cross-cultural service robots, the "handing over" action needs to meet both ergonomic constraints (NCR) and cultural etiquette requirements (CDR).

[0107] Through a dual-domain processing flow, the verb interface ensures seamless transfer of knowledge between different domains.

[0108] 4. Verb Causal Chain Tracing

[0109] The verb interface can clearly describe the evolution of complex behaviors by establishing a causal chain of verbs. This causal chain is not only applicable to long-range reasoning in a single domain, but can also be used for cross-domain knowledge transfer. For example:

[0110] In the field of financial trading, the causal chain of the verb “go long” can be extended to other scenarios involving the flow of value (such as supply chain management or energy trading).

[0111] In the field of industrial manufacturing, the causal chain of the verb “cut” can be extended from the physical cutting process to material modeling in virtual simulation.

[0112] By tracing the causal chain of verbs, the verb interface is able to transfer knowledge from one domain to another while retaining its core logic and causal relationships.

[0113] 5. Verb-noun joint modeling

[0114] The verb interface enhances the dynamic relevance and adaptability of knowledge through verb-noun joint modeling. Specifically,

[0115] Verbs are noun-dependent: The meaning and function of a verb are concretized by nouns (e.g., subject, frame of reference, property change). For example, the verb "cut" depends on the knife (subject), the wood (frame of reference), and the change in molecular structure (property change).

[0116] Nouns rely on verbs: Nouns form a dynamic network of relationships under the influence of verbs, giving dynamic meaning to static entities. For example, the noun "knife" acquires dynamic meaning through the verb "cut."

[0117] Through this joint modeling approach, the verb interface can transform static nouns into dynamic relational networks, thereby providing richer semantic support for cross-domain migration.

[0118] Beneficial effects:

[0119] 1. Unified modeling capabilities

[0120] Verb theory constructs a "verb-noun dynamic network" by clarifying the dynamic association between the four-tuple structure of verbs (subject, reference system, spatiotemporal dimension, and property change) and nouns, thereby enhancing the dynamic association and adaptability of knowledge. This unified modeling capability enables knowledge from different fields to be expressed in a consistent manner, thus providing a basis for cross-domain migration. Adaptation to different scenarios can be achieved by flexibly adjusting specific parameters. For example, the action of "grasping" can be represented as [robotic arm → part → 3D space → contact force] in an industrial assembly scenario, while in a medical surgery scenario it can be represented as [surgical instrument → tissue → time accuracy → tension control error]. Although the specific parameters are different, the core structure remains the same.

[0121] 2. Dual-domain unified computing architecture

[0122] Verb theory achieves unified modeling of the objective reality domain and the subjective cognitive domain by simultaneously processing the non-cognitive domain (NCR) and the cognitive dependency domain (CDR). This dual-domain processing flow significantly improves the adaptability of the system, enabling the system to flexibly switch between physical rules and cultural backgrounds. For example, in the field of autonomous driving, "avoidance" actions need to verify whether the change in trajectory curvature conforms to the physical equation of motion (NCR), while also considering emotional coupling decision-making (CDR). In the field of cross-cultural service robots, "handing over" actions need to meet ergonomic constraints (NCR) and cultural etiquette requirements (CDR).

[0123] 3. Verb interface standardization

[0124] By defining a four-tuple structure for verbs, verb theory provides a standardized interface that enables knowledge from different domains to be expressed and transferred through the same framework. This standardized form not only simplifies the migration of cross-domain knowledge but also improves the scalability and compatibility of the system. For example, the representation of the verb "cut" in industrial manufacturing and its semantic parsing in natural language can be seamlessly integrated through the same framework. The verb interface can decompose complex dynamic behaviors into a set of clear elements, providing a foundation for the cross-domain migration of knowledge.

[0125] 4. Verb Causal Chain Tracing

[0126] Verb theory can clearly describe the evolution of complex behaviors by establishing causal chains of verbs. This causal chain is not only applicable to long-range reasoning in a single field, but can also be used for cross-domain knowledge transfer. For example, in the field of financial trading, the causal chain of the verb "go long" can be extended to other scenarios involving value flow (such as supply chain management or energy trading). In the field of industrial manufacturing, the causal chain of the verb "cut" can be extended from the physical cutting process to material modeling in virtual simulation.

[0127] 5. Cognitive paradigm innovation

[0128] Breaking through the traditional AI cognitive model centered on noun entities, a new paradigm of "verb-noun dynamic networks" is established. This innovation fundamentally shifts the cognitive model from static node-based associations to modeling dynamic process units. For example, static node-based associations (such as hospital → doctor → treatment → patient) are transformed into dynamic process units (such as [doctor → patient → operating room → organ removal]). This is more closely aligned with real-world application scenarios, especially demonstrating significant advantages in describing dynamic behaviors and their evolution.

[0129] 6. Explicit Spatiotemporal Modeling

[0130] Verb theory quantifies and encodes the spatiotemporal dimensions as essential elements of verbs, resolving the ambiguity of temporal relationships in traditional methods. This explicit modeling approach ensures the accuracy of the temporal and spatial scope of actions. For example, in autonomous driving scenarios, the temporal and spatial scope of actions are clearly defined. For example, an "avoid" action requires a change in trajectory curvature within 0.5 seconds. In the high-frequency trading verb map, the spatiotemporal dimension of the "go long" verb is the price volatility within a specific time window.

[0131] 7.Quantization expression ability

[0132] Verb theory has achieved numerous technological innovations in quantum representation, including ground state vectors representing action eigenstates, entangled states representing cross-cultural action associations, and quantum gate operations corresponding to action conversion rules. These technologies significantly improve the ability to accurately model and efficiently convert actions. For example, ① Ground state vectors represent action eigenstates: The four-tuple structure of a verb is mapped into quantum Hilbert space, and the ground state vectors represent the action eigenstates. For example, the state of the verb "cut" includes the subject (the tool), the reference frame (the wood), the spacetime dimension (linear time), and the property change (a change in molecular structure). ② Entangled states represent cross-cultural action associations: Quantum entangled states are used to describe action associations in different cultural contexts. For example, the differences between Chinese and Western bowing can be represented by entangled states. ③ Quantum gate operations correspond to action conversion rules: The action conversion rules are mapped to quantum gate operations. For example, the energy level transition from "walk to run" can be realized using a specific quantum gate.

[0133] 8. Cross-cultural adaptability

[0134] By leveraging hypergraph neural networks and quantum entanglement, verb theory significantly improves the system's adaptability across different cultural contexts, significantly reducing misjudgment rates. For example, in the field of cross-cultural service robotics, the specific form of the "handing over" action can be adjusted to suit different cultural contexts. For example, in a Japanese model, it might be expressed as both hands and a 15-degree bow, while in a Middle Eastern model, it might be expressed as the right hand and avoiding eye contact.

[0135] In summary, through the above eight technical innovations, verb theory has achieved comprehensive analysis and optimization of dynamic behaviors, providing new theoretical and technical support for the development of strong artificial intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0136] Figure 1 It is a diagram of the verb tetragram structure;

[0137] Figure 2 This is a diagram of the dual-domain unified computing architecture. DETAILED DESCRIPTION

[0138] In order to better illustrate the technical solution and application of the present invention, the present invention is described in detail below through two specific embodiments. It should be noted that the following embodiments are only exemplary descriptions of the present invention, and the protection scope of the present invention is not limited thereto.

[0139] Example 1: Modeling of “avoidance” actions in the field of autonomous driving

[0140] 1. Background

[0141] In autonomous driving scenarios, "avoidance" refers to the vehicle's trajectory adjustment to avoid collision with an obstacle. This action requires precise control in both time and space, while also considering the physical equations of motion (NCR) and the coupled decision decision (CDR). This example comprehensively analyzes and optimizes "avoidance" actions based on verb theory.

[0142] 2. Definition of Verb Quadruple Structure

[0143] 1. Subject

[0144] Definition: The entity that performs the "avoidance" action is a vehicle.

[0145] Technical implementation: Use vehicle-mounted sensors (such as lidar and cameras) to perceive the surrounding environment in real time and determine the vehicle as the subject of the "avoidance" action.

[0146] Practical application: For example, in a highway scenario, the vehicle body may need to take evasive action based on an obstacle that suddenly appears in front of it (such as a fallen tire).

[0147] 2.Reference (reference system)

[0148] Definition: The object of the action is the obstacle.

[0149] Technical Implementation: Identify and locate the position and size of obstacles through an environmental perception system (such as a deep learning model). For example, use the YOLO (You Only Look Once) object detection algorithm to identify obstacle types (such as pedestrians, other vehicles, or stationary objects).

[0150] Practical application: In urban driving scenarios, obstacles may be pedestrians or bicycles that suddenly enter the lane.

[0151] 3. Time / Space

[0152] Definition: Describes the time and space range of the action, and sets the trajectory curvature change to complete within 0.5 seconds.

[0153] Technical implementation:

[0154] Time dimension: Determine the time window required for action by calculating the relative speed and distance between the vehicle and the obstacle in real time.

[0155] Spatial dimension: Use high-precision maps and navigation systems to plan the vehicle's avoidance path in three-dimensional space.

[0156] Practical application: For example, on a highway, a vehicle needs to switch from the current lane to the adjacent lane within 0.5 seconds to avoid a vehicle in front that suddenly brakes.

[0157] 4. Property (property change)

[0158] Definition: The change in state or property caused by an action is called a change in trajectory curvature.

[0159] Technical implementation: The change in trajectory curvature is calculated through the vehicle dynamics model and ensured to comply with physical constraints (such as maximum lateral acceleration limit).

[0160] Practical application: For example, in a curve scenario, the vehicle needs to adjust the steering angle to maintain a safe driving trajectory.

[0161] 3. Technical Implementation

[0162] 1. Dual-domain unified computing architecture

[0163] (1) Physical domain verification

[0164] Theoretical basis: Based on the objective reality in the non-cognitive domain (NCR), including basic existences such as the basic physical layer, formalized system and pre-observation ontology.

[0165] Technical implementation:

[0166] Quantum Monte Carlo simulation is used to verify whether the change in vehicle trajectory curvature conforms to the physical equations of motion.

[0167] Specific steps:

[0168] ① Input vehicle parameters (such as mass, wheelbase, maximum lateral acceleration, etc.).

[0169] ②Simulate the physical feasibility of different avoidance paths.

[0170] ③Output the optimal path.

[0171] Practical application: For example, in emergency obstacle avoidance, verify whether the vehicle can complete the evasive action without exceeding the tire grip.

[0172] (2) Cognitive Domain Reasoning

[0173] Theoretical basis: Based on the knowledge system in the cognitive dependency domain (CDR), including the individual layer (S), group layer (OS) and social layer (IS).

[0174] Technical implementation:

[0175] Capturing cultural context information of actions using hypergraph neural networks.

[0176] Specific steps:

[0177] ① Collect behavioral data of drivers from different cultural backgrounds.

[0178] ② Train a hypergraph neural network to identify the impact of cultural differences on avoidance actions.

[0179] ③Dynamically adjust avoidance strategies to adapt to different cultural backgrounds.

[0180] Practical application: For example, in the Chinese model, “avoidance” may manifest as a more aggressive turn, while in the European model it may focus more on stability.

[0181] 2. Innovation in Quantum Expression

[0182] (1) The base state vector represents the action eigenstate

[0183] Theoretical basis: Map the four-tuple structure of the verb into the quantum Hilbert space and use the ground state vector to represent the eigenstate of the action.

[0184] Technical implementation:

[0185] The "avoid" verb is mapped into quantum Hilbert space, and the eigenstate of the action is represented by the ground state vector.

[0186] Specific steps:

[0187] ① Define the ground state vector |avoid>, which includes the subject (vehicle), reference frame (obstacle), space-time dimension (0.5 seconds), and property change (change in trajectory curvature).

[0188] ② Use the principle of quantum state superposition to construct the state space of action.

[0189] Practical Application: For example, in the optimization of flexible assembly actions, the state of the “avoid” verb can be represented as a linear combination of |avoid success> and |failure> through quantum state superposition.

[0190] (2) Entangled states represent cross-cultural action associations

[0191] Theoretical basis: Using quantum entangled states to describe action associations in different cultural contexts.

[0192] Technical implementation:

[0193] Use quantum entanglement to describe the different understandings of "avoidance" maneuvers by Chinese and Western drivers.

[0194] Specific steps:

[0195] ①Define entangled states |avoid_China> and |avoid_Europe>.

[0196] ② Dynamically adjust the avoidance strategy by measuring the correlation strength of the entangled state.

[0197] Practical applications: For example, in the field of cross-cultural service robots, the specific form of "avoidance" actions can be adjusted according to different cultural backgrounds.

[0198] (3) Quantum gate operation corresponding action conversion rules

[0199] Theoretical basis: Mapping the action conversion rules into quantum gate operations.

[0200] Technical implementation:

[0201] Map the “go straight → avoid” energy level transition to a specific quantum gate operation.

[0202] Specific steps:

[0203] ① Define the quantum gate U_avoid, which represents the transition from the straight-forward state to the avoidance state.

[0204] ② Apply quantum gate operations to achieve optimal path selection.

[0205] Practical applications: For example, in an autonomous driving emergency obstacle avoidance decision-making system, the trajectory change of the "avoid" verb can achieve optimal path selection through a series of quantum gate operations.

[0206] IV. Practical Application

[0207] 1. Verb-noun joint modeling

[0208] Verbs depend on nouns:

[0209] Clarify the four-tuple structure of the "avoid" verb and transform the abstract verb into a specific action description.

[0210] Specific steps:

[0211] ① Determine the subject (vehicle), reference system (obstacle), space-time dimension (0.5 seconds) and property change (change in trajectory curvature).

[0212] ② Establish a verb-noun association matrix.

[0213] Nouns rely on verbs:

[0214] Through the action of verbs, static nouns are transformed into dynamic relational networks.

[0215] Specific steps:

[0216] ①Define the dynamic relationship between the vehicle and the obstacle.

[0217] ②Realize real-time trajectory calculation and adjustment.

[0218] 2. System performance improvement

[0219] Error reduction: By clarifying the quadruple structure of verbs and the dynamic association of nouns, the error rate of actions is significantly reduced.

[0220] Improved efficiency: By establishing a verb-noun association matrix, real-time calculation of trajectory changes is achieved.

[0221] Enhanced security: Through a dual-domain unified computing architecture, actions are ensured to comply with physical laws and cultural backgrounds.

[0222] In summary, this embodiment comprehensively analyzes and optimizes the "avoidance" action in the field of autonomous driving through verb theory, and provides detailed technical implementation steps and practical application scenarios, making it easier for technical personnel in related fields to understand and operate.

[0223] Example 2: Modeling of “resection” action in medical surgery

[0224] 1. Background

[0225] In medical surgery, the "resection" action refers to the act of cutting and removing diseased tissue with a surgical instrument. This action requires precise temporal and spatial control, while also considering parameters such as the instrument's force threshold and the elastic modulus of biological tissue. This example comprehensively analyzes and optimizes the "resection" action based on verb theory.

[0226] 2. Definition of Verb Quadruple Structure

[0227] 1. Subject

[0228] Definition: The entity that performs the "cutting" action is a surgical instrument.

[0229] Technical implementation:

[0230] The surgical instrument performs the "cutting" action through a robotic control system or manual operation by the surgeon.

[0231] Specific steps:

[0232] ① Use high-precision sensors (such as force sensors and displacement sensors) to monitor the status of surgical instruments in real time.

[0233] ② Ensure that the surgical instrument, as the main body of the "resection" action, can operate stably.

[0234] Practical application: For example, in laparoscopic surgery, the surgical instrument may be an electric scalpel or an ultrasonic scalpel.

[0235] 2.Reference (reference system)

[0236] Definition: The object of the action is the tumor boundary.

[0237] Technical implementation:

[0238] Identify and locate tumor boundaries through medical imaging processing systems (such as CT, MRI).

[0239] Specific steps:

[0240] ① Use deep learning models (such as U-Net) to segment the tumor area.

[0241] ② Determine the tumor boundary as the reference system for the "resection" action.

[0242] Practical Application: During brain surgery, tumor boundaries may be precisely located using magnetic resonance imaging (MRI).

[0243] 3. Time / Space

[0244] Definition: Describes the time and space range of an action, set as the operation accuracy within a specific time window.

[0245] Technical implementation:

[0246] Time dimension: By calculating the relative position and velocity between the surgical instrument and the tumor boundary in real time, the time window required for the action is determined.

[0247] Spatial dimension: Use a three-dimensional navigation system to plan the resection path of surgical instruments in three-dimensional space.

[0248] Practical application: For example, in minimally invasive surgery, surgical instruments need to complete the resection operation from the edge of the tumor to the center within a limited time.

[0249] 4. Property (property change)

[0250] Definition: A change in state or quality caused by an action such as the removal of an organ.

[0251] Technical implementation:

[0252] The biomechanical model is used to calculate the deformation and damage of biological tissues during the resection process and ensure that they meet safety standards.

[0253] Specific steps:

[0254] ① Define the elastic modulus and strength limit of biological tissues.

[0255] ②Calculate the stress and strain generated during the removal process.

[0256] Practical application: For example, in liver surgery, the resection process needs to ensure the functional integrity of the remaining liver tissue.

[0257] 3. Technical Implementation

[0258] 1. Dual-domain unified computing architecture

[0259] (1) Physical domain verification

[0260] Theoretical basis: Based on the objective reality in the non-cognitive domain (NCR), including basic existences such as the basic physical layer, formalized system and pre-observation ontology.

[0261] Technical implementation:

[0262] Quantum Monte Carlo simulation is used to verify whether the force acting on the device meets the mechanical equilibrium conditions.

[0263] Specific steps:

[0264] ① Input surgical instrument parameters (such as mass, stiffness, maximum force, etc.).

[0265] ②Simulate the physical feasibility of different resection paths.

[0266] ③Output the optimal path.

[0267] Practical application: For example, in orthopedic surgery, verify whether the surgical instrument can complete the resection without exceeding the bearing capacity of the bone.

[0268] (2) Cognitive Domain Reasoning

[0269] Theoretical basis: Based on the knowledge system in the cognitive dependency domain (CDR), including the individual layer (S), group layer (OS) and social layer (IS).

[0270] Technical implementation:

[0271] Capturing cultural context information of actions using hypergraph neural networks.

[0272] Specific steps:

[0273] ① Collect behavioral data on surgeons in different medical environments (such as developed countries and developing countries).

[0274] ② Train a hypergraph neural network to identify the impact of cultural differences on resection actions.

[0275] ③Dynamically adjust resection strategies to adapt to different medical environments.

[0276] Practical application: For example, in a developed country model, “resection” may focus more on minimally invasiveness, while in a developing country model it may focus more on cost-effectiveness.

[0277] 2. Innovation in Quantum Expression

[0278] (1) The base state vector represents the action eigenstate

[0279] Theoretical basis: Map the four-tuple structure of the verb into the quantum Hilbert space and use the ground state vector to represent the eigenstate of the action.

[0280] Technical implementation:

[0281] The verb "remove" is mapped into quantum Hilbert space, and the eigenstate of the action is represented by the ground state vector.

[0282] Specific steps:

[0283] ① Define the ground state vector |resection>, which includes the subject (surgical instrument), the reference frame (tumor boundary), the space-time dimension (specific time window), and the property change (organ resection).

[0284] ② Use the principle of quantum state superposition to construct the state space of action.

[0285] Practical application: For example, in the high-frequency trading verb map, the state of the verb "resection" can be represented by quantum state superposition as a linear combination of |resection success> and |failure>.

[0286] (2) Entangled states represent cross-cultural action associations

[0287] Theoretical basis: Using quantum entangled states to describe action associations in different cultural contexts.

[0288] Technical implementation:

[0289] Using quantum entangled states to describe the different understandings of the "resection" action by surgeons in developed and developing countries.

[0290] Specific steps:

[0291] ① Define the entangled states |remove_developed_country> and |remove_developing_country>.

[0292] ② Dynamically adjust the excision strategy by measuring the correlation strength of the entangled state.

[0293] Practical applications: For example, in the field of cross-cultural service robots, the specific form of the "resection" action can be adjusted according to different medical environments.

[0294] (3) Quantum gate operation corresponding action conversion rules

[0295] Theoretical basis: Mapping the action conversion rules into quantum gate operations.

[0296] Technical implementation:

[0297] The energy level transition of "positioning→removal" is mapped to a specific quantum gate operation.

[0298] Specific steps:

[0299] ① Define quantum gate U_cut, which represents the transition from the localized state to the cut state.

[0300] ② Apply quantum gate operations to achieve optimal path selection.

[0301] Practical applications: For example, in the field of medical diagnosis, in the risk entropy calculation of the verb "resection", quantum gate operations are used to dynamically adjust the threshold range of the instrument's force, thereby reducing the risk of surgical actions.

[0302] IV. Practical Application

[0303] 1. Verb-noun joint modeling

[0304] Verbs depend on nouns:

[0305] Clarify the four-tuple structure of the verb "cut off" and transform the abstract verb into a specific action description.

[0306] Specific steps:

[0307] ① Determine the subject (surgical instrument), reference frame (tumor boundary), spatiotemporal dimension (specific time window), and property change (organ resection).

[0308] ② Establish a verb-noun association matrix.

[0309] Nouns rely on verbs:

[0310] Through the action of verbs, static nouns are transformed into dynamic relational networks.

[0311] Specific steps:

[0312] ① Define the dynamic relationship between surgical instruments and tumor boundaries.

[0313] ②Realize real-time resection path calculation and adjustment.

[0314] 2. System performance improvement

[0315] Error reduction: By clarifying the quadruple structure of verbs and the dynamic association of nouns, the error rate of actions is significantly reduced.

[0316] Improved efficiency: By establishing a verb-noun association matrix, real-time calculation of the resection path is achieved.

[0317] Enhanced security: Through a dual-domain unified computing architecture, actions are ensured to comply with physical laws and medical environment requirements.

[0318] In summary, this embodiment comprehensively analyzes and optimizes the "resection" action in the field of medical surgery through verb theory, and provides detailed technical implementation steps and practical application scenarios, making it easier for technical personnel in related fields to understand and operate.

[0319] The above content is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the field within the technical scope disclosed by the present invention should be covered by the scope of protection of the present invention.

Claims

1. A method for constructing a dynamic process network based on verb-noun joint modeling, characterized in that: The following steps are involved: The verb is defined as a four-tuple structure (subject, reference frame, spatiotemporal dimension, and property change), where each element is a noun or a nominal entity; Construct a "verb-noun dynamic network" to analyze and optimize dynamic behaviors through the interaction between the verb quadruple structure and related nouns.

2. The method according to claim 1, characterized in that The verb definition further includes: The subject of the movement is the entity or object that performs the action; The reference frame is the object or environment on which the action is taken; Clarify the time and space dimensions (Time / Space) to describe the time and space range of the action; A property change indicates a change in state or property caused by an action.

3. The method according to claim 1, characterized in that The verb-noun joint modeling includes the following: Verbs are dependent on nouns: the meaning and function of verbs need to be concretized by nouns (such as subject, reference system, and property change); Nouns rely on verbs: nouns form a dynamic relationship network under the action of verbs, giving static entities dynamic meaning; Joint modeling: By establishing a verb-noun association matrix, comprehensive analysis and optimization of dynamic behaviors are achieved.

4. The method according to any one of claims 1 to 3, characterized in that It also includes a dual-domain unified computing architecture, specifically including: Physical domain verification: Quantum Monte Carlo simulation is used to check the physical feasibility of actions and ensure that all dynamic behaviors conform to the laws of nature. Cognitive domain reasoning: Using hypergraph neural networks to parse the meaning of actions in cultural contexts and enhance adaptability to subjective cognitive domains.

5. The method according to claim 4, characterized in that The physical domain verification further comprises: Real-time verification of physical constraints based on objective reality in the non-cognitive domain (NCR); Tensor cores are used to verify the physical feasibility of the action, ensuring that the action complies with the physical equations of motion and mechanical equilibrium conditions.

6. The method according to claim 4, characterized in that The cognitive domain reasoning further includes: Based on the knowledge system in the cognitive dependency domain (CDR), it includes the individual layer (S), the group layer (OS) and the social layer (IS); The cultural background information of the action is captured through the hypergraph neural network, and the specific form of the action is dynamically adjusted under different cultural backgrounds.

7. The method according to any one of claims 1 to 6, characterized in that It also includes innovations in quantum expression, including: The ground state vector represents the eigenstate of the action: the four-tuple structure of the verb is mapped into the quantum Hilbert space, and the ground state vector is used to represent the eigenstate of the action; Entangled states represent cross-cultural action correlations: using quantum entangled states to describe action correlations in different cultural contexts; Quantum gate operations correspond to action conversion rules: map the action conversion rules to quantum gate operations.

8. The method according to claim 7, characterized in that The base state vector represents the action eigenstate further comprising: Map the four-tuple structure of the verb (subject, reference frame, space-time dimension, property change) into quantum Hilbert space; The state space of action is constructed through the principle of quantum state superposition.

9. The method according to claim 7, characterized in that The entangled state representation of cross-cultural action association further includes: Using quantum entangled states to describe the association of actions in different cultural contexts; In the field of cross-cultural service robots, the hardware supports the cognitive dependency domain (CDR) reasoning engine to run cultural rules with ultra-low power consumption.

10. The method according to claim 7, characterized in that The quantum gate operation corresponding action conversion rules further include: Map the action conversion rules into quantum gate operations; In the autonomous driving emergency obstacle avoidance decision-making system, the optimal path selection is achieved through a series of quantum gate operations.

11. The method according to any one of claims 1 to 10, characterized in that It also includes a verb interface to achieve automatic knowledge transfer, including: Unified modeling capabilities: By clarifying the four-tuple structure of verbs (subject, reference frame, spatiotemporal dimension, and property change) and the dynamic association of nouns, a standardized knowledge expression and transfer framework is provided; Parameterized adjustment: Adaptation to different fields is achieved by adjusting specific parameters in the verb quadruple structure; Dual-domain unified computing architecture: Simultaneously processes the non-cognitive domain (NCR) and cognitive dependency domain (CDR), flexibly switching between the objective reality domain and the subjective cognitive domain; Verb causal chain tracing: By establishing verb causal chains, the evolution process of complex behaviors can be clearly described; Verb-noun joint modeling: Enhancing the dynamic relevance and adaptability of knowledge.

12. The method according to claim 11, characterized in that The verb interface realizes automatic knowledge migration further comprising: In the field of industrial robotics, the four-tuple structure of the "grasp" verb is [robotic arm → part → 3D space → contact force]; In the field of medical surgery, the four-tuple structure of the verb "grasp" is [surgical instrument → tissue → time accuracy → tension control error].

13. The method according to any one of claims 1 to 12, characterized in that The following benefits are also included: Unified modeling capabilities: Enhance the dynamic relevance and adaptability of knowledge; Dual-domain unified computing architecture: Improves system adaptability and enables flexible switching between physical rules and cultural contexts; Verb interface standardization: simplifies the migration of cross-domain knowledge and improves the scalability and compatibility of the system; Verb Causal Chain Tracing: This technology clearly describes the evolution of complex behaviors and is suitable for long-term reasoning in a single domain and cross-domain knowledge transfer. Cognitive Paradigm Innovation: This technology breaks through the traditional AI cognitive model centered on noun entities and establishes a new paradigm of "verb-noun dynamic network." Explicit spatiotemporal modeling: This solves the problem of ambiguous temporal relationships in traditional methods and ensures the accuracy of the temporal and spatial ranges of actions. Quantized expression capability: This significantly improves the ability to accurately model and efficiently convert actions. Cross-cultural adaptability: The system's adaptability in different cultural contexts is significantly improved, and the misjudgment rate is greatly reduced.

Images