Intelligent system paradigm operating system and implementation method thereof

Abstract

The invention discloses an intelligent system paradigm operating system and an implementation method thereof. The system realizes six-stage system-level intelligent circulation of perception, cognition, decision making, execution, feedback and learning in an operating system layer. According to the system, a three-domain collaborative system is formed by a perception driving domain, a cognitive decision domain and an interaction expression domain, cross-domain communication, encryption verification and state synchronization are achieved through a unified data protocol and a trust bridge mechanism, and operation consistency, safety and verifiability of the system in a multi-terminal and multi-node environment are ensured. The system has adaptive optimization and continuous learning capabilities, can automatically adjust operation parameters and execution strategies according to task results and environmental changes, and loads an external function unit through a hierarchical expansion interface to realize function expansion and ecological level evolution. According to the invention, a unified structure basis and a system normal form are provided for a future artificial intelligence operating system, a cognitive computing platform and an autonomous evolution type intelligent system.

Description

Technical Field

[0001] This invention relates to the fields of artificial intelligence and computer operating system technology, specifically to an Intelligent System Paradigm Operating System (IS-OS) and its implementation method. Background Technology

[0002] With the rapid development of artificial intelligence (AI) and large-scale modeling technology, intelligent computing capabilities have made significant progress at the application layer. However, existing computer operating systems (such as Windows, Linux, and Android) are essentially still "passive response systems," whose core functions are limited to resource management, process scheduling, and hardware abstraction, lacking system-level proactive cognition and decision-making capabilities.

[0003] In the existing technological system, intelligent functions are provided by applications (such as voice assistants and image recognition software) or external AI models, with the operating system merely acting as a passive intermediary for resource requests. The typical workflow is "user → application → system," meaning the application requests resources from the operating system to complete a specific function. This architecture has the following inherent drawbacks:

[0004] (1) Lack of system-level intelligent loop: Intelligent behavior depends on the invocation of external applications or models. The operating system itself does not have a continuous closed loop of perception, cognition, decision-making and learning, and cannot understand the task intention, carry out autonomous planning and long-term optimization from the system level.

[0005] (2) The trust and security mechanisms are separated: system security (such as user accounts), application permissions and AI model access control are independent of each other, lacking a unified identity verification and trust transmission mechanism that runs through the whole system, making it difficult to ensure the security consistency of cross-module and cross-terminal task execution.

[0006] (3) Insufficient interaction and ecosystem scalability: Device collaboration and plug-in ecosystem are highly dependent on specific platform interfaces. The operating system lacks an abstract and universal capability extension framework, making it difficult to seamlessly integrate diverse external functional units, which hinders the evolution of the intelligent ecosystem.

[0007] (4) Lack of system self-optimization capability: Traditional operating systems cannot adjust and optimize themselves based on historical execution feedback. Parameters and strategies are usually statically configured and do not have the ability to continuously learn from running experience and improve their behavioral logic.

[0008] Therefore, existing technologies urgently need a new operating system paradigm that possesses system-level intelligent attributes, enabling it to proactively perceive the environment, understand tasks, make decisions, execute actions, and learn from them, thereby achieving a fundamental shift from "passive response" to "proactive intelligence". Summary of the Invention

[0009] This invention proposes an intelligent system paradigm operating system and its implementation method, which enables the operating system as a whole to possess intelligent system attributes. By building and running a complete intelligent loop at the system kernel level, it achieves autonomous perception, cognition, decision-making, execution, feedback and learning.

[0010] To address the aforementioned technical problems, this invention provides the following technical solution: This solution provides an Intelligent System Paradigm Operating System (IS-OS), which possesses intelligent system attributes. The system has a built-in and continuously running system-level intelligent loop at the operating system layer, consisting of six stages: Perception, Cognition, Decision, Action, Feedback, and Learning. This system includes the following three core functional domains:

[0011] Perception-driven domain (L1): As the physical and execution foundation layer of the system, it is responsible for multimodal signal acquisition, hardware and software resource scheduling, issuing and running specific task execution instructions, and collecting task execution results and system status data as raw feedback. It can be implemented by the kernel module of the operating system, device driver framework and resource manager.

[0012] Cognitive Decision Domain (L2): As the core intelligent layer of the system, it is responsible for semantic understanding, contextual analysis, logical reasoning, and task planning of the perceptual information input from L1, generating specific decision-making schemes. Simultaneously, it is responsible for analyzing the feedback data collected by L1, updating the cognitive model and decision-making strategies through built-in learning algorithms to achieve self-optimization. It can be implemented as one or a group of intelligent engines within the operating system kernel, such as integrating a lightweight neural network inference engine, a rule engine, and a knowledge graph.

[0013] Interactive Expression Domain (L3): As the interface layer for the interaction between the system and the internal and external environment, it is responsible for receiving user input, external system instructions or environmental signals, and transmitting them into the system after standardization; at the same time, it is responsible for outputting the system's internal decision results and status information to the user or external entities in a multimodal form. It can be built based on the operating system's input / output subsystem, graphics services and network communication stack.

[0014] Furthermore, the three functional domains are connected and coordinated through two core mechanisms:

[0015] Unified Data Protocol (UDP / UDF): Defines the message format, data structure, authentication fields, and status coding rules for cross-domain communication, ensuring the syntactic consistency of data flow between L1, L2, and L3, including but not limited to the data layer, authentication layer, and trust layer structures;

[0016] Trust Bridge Mechanism (TBM): As a security pillar spanning the three domains, it performs mandatory identity authentication, permission verification, message integrity protection, and encrypted transmission for all cross-domain communications, and maintains a traceable trust chain to ensure that the operation of each link in the intelligent loop is authorized and verifiable.

[0017] Furthermore, the perception-driven domain (L1) includes, but is not limited to, signal acquisition, resource scheduling, task execution and feedback acquisition modules, which are responsible for realizing the logical closure of the perception, execution and feedback stages in the system loop, and constitute the physical execution and perception foundation layer of the system.

[0018] Furthermore, the cognitive decision domain (L2) includes, but is not limited to, semantic understanding, logical reasoning, task planning, trust assessment and learning optimization modules. This domain generates decision plans based on feedback from L1 and input from L3, and achieves resource coordination and strategy transfer through UDP / UDF and TBM, forming the core cognitive and learning layer of the system loop.

[0019] Furthermore, the interactive expression domain (L3) includes, but is not limited to, input acquisition, output expression, multimodal interaction and external communication modules, which are used to connect users, environment and external systems to form a closed-loop interactive mechanism of input-execution-feedback-output.

[0020] The system center serves as a unified hub connecting various domains and supports the addition of new scalable system domains (Lx) to accommodate future system-level functional expansions. External functional units (EFU / CU) connect to the system kernel via extension interfaces (LEI), providing the system with open capability access.

[0021] This invention proposes an implementation method for an intelligent system paradigm operating system. The operation process of this system is as follows: Figure 1-5 As shown, the interactive expression domain (L3), the perception-driven domain (L1), and the cognitive decision-making domain (L2) form a closed loop according to the collaborative rhythm of "L3→L1→L2→L2→L1→L1→L2→L3". The system-level intelligent loop implemented at the operating system layer includes the following steps:

[0022] Step 1: L3 input stage. The interactive expression domain (L3) receives the raw input signal through the input device or network interface, performs preprocessing and standardization, verifies it through the Trust Bridge Mechanism (TBM), encapsulates it according to the unified data protocol (UDP / UDF), and sends it to the perception driving domain (L1).

[0023] Step 2: In the L1 perception phase, the perception driver domain (L1) receives the standardized data packets from L3, detects the current resource status of the system through the status monitoring module, and generates a system perception report based on the input signals. The perception report is encrypted by TBM and transmitted to L2. At the same time, L1 initializes the task execution resource pool to prepare for subsequent task execution.

[0024] Step 3: In the L2 cognitive stage, the semantic understanding module of the cognitive decision domain (L2) analyzes the input signals in the perception report, identifies user intent or environmental needs, and generates a task semantic model; the logical reasoning module combines the system's historical state data and trust assessment results to determine the resources and external units required for task execution.

[0025] Step 4: L2 Decision-Making Phase. The L2 task planning module formulates specific execution strategies based on semantic models and resource assessments, and generates a decision data packet containing execution instructions and authorization information. The decision data packet is then transmitted to L1 after being signed by TBM.

[0026] Step 5: During the L1 execution phase, the L1 resource scheduling module allocates system resources according to the decision instructions and calls the specified external functional unit (EFU) through the extended interface (LEI) to control the EFU to execute specific tasks; during task execution, L1 collects execution status data in real time.

[0027] Step 6: L1 Feedback Phase. After the task is completed, the L1 feedback acquisition module integrates the execution result data and system resource consumption data to generate a feedback report. The feedback report is transmitted to L2 after being standardized by UDP / UDF and encrypted by TBM.

[0028] Step 7: In the L2 learning phase, the L2 learning optimization module analyzes the feedback report. If the task execution meets the target, the historical task database is updated to strengthen the corresponding decision model; if it does not meet the target, the decision strategy is optimized through reinforcement learning algorithm, and the updated model parameters are stored in the system knowledge base.

[0029] Step 8: L3 output stage. L2 transmits the optimization results and execution feedback to L3. The L3 output encoding module converts the information into a multimodal form and feeds it back to the user or external system, completing one intelligent loop.

[0030] All data flows in the above process must be verified by the Trust Bridge Mechanism (TBM) under the premise of conforming to the unified data protocol (UDP / UDF). The system completes a full six-stage intelligent cycle under the cooperation of three domains: perception → cognition → decision-making → execution → feedback → learning.

[0031] Innovations compared to traditional systems:

[0032] Unlike the traditional one-way call structure of "user → application → system", this system uses a multi-layer loop logic of "L3–L1–L2–L1–L2–L3" to realize the system layer's proactive understanding, feedback and optimization capabilities, enabling the operating system to have long-term learning, self-adaptation and feedback consistency characteristics, and constructing an intelligent system paradigm that can run in a self-looping manner on any software and hardware platform.

[0033] The present invention proposes an intelligent system paradigm operating system and its implementation method. The beneficial effects achieved by adopting the above structure are as follows:

[0034] (1) Achieved system-level native intelligence: By directly integrating L1, L2, and L3 domains and intelligent loop logic into the operating system kernel layer, intelligence becomes an inherent attribute of the operating system, rather than an externally added function. This solves the problem that traditional system intelligence relies on the application layer and cannot achieve system-level linkage, enabling deeper resource collaboration and global optimization.

[0035] (2) A lightweight and scalable intelligent architecture has been constructed: the three-domain decoupled design allows each domain to evolve and be deployed independently. For example, the L2 cognitive engine can be replaced and upgraded without affecting the L1 execution framework. Through the hierarchical extension interface (LEI) and unified trust verification (TBM), various external functional units (EFU) can be securely and conveniently connected, which greatly enhances the system's ecological compatibility and functional scalability, and overcomes the limitations of traditional systems that are closed ecosystems and difficult to expand.

[0036] (3) Ensures security and trust consistency across domains and nodes: The unified Trust Bridge (TBM) mechanism and data protocol (UDP / UDF) provide end-to-end security for all interactions within the intelligent loop. Whether it is communication between three domains within the same device or collaboration across nodes in a distributed deployment, it can ensure trusted identity, complete messages, and clear authorization, thus building a system-level Trust Network (STN) and solving the problem of fragmented security mechanisms in traditional architectures.

[0037] (4) It endows the system with continuous self-optimization and evolution capabilities: The learning stage inherent in L2 enables the system to automatically adjust its strategy based on the feedback of each execution. For example, by analyzing the success rate and time consumption of task execution, the system can automatically optimize the resource scheduling algorithm or decision threshold. This gives the system long-term adaptive capabilities and enables it to continuously improve its performance as it is used, breaking through the bottleneck of traditional operating system parameters being fixed and lacking learning capabilities.

[0038] (5) Supports flexible cross-platform deployment: Due to the abstract nature of the core architecture, the system described in this invention can be adapted to a variety of hardware platforms. The three domains can be centrally deployed on a single device or distributed on different nodes in the cloud, edge, and terminal. By using UDP / UDF and TBM to maintain the continuity of the loop logic and state synchronization, seamless cross-platform migration and collaboration of intelligent capabilities are achieved. Attached Figure Description

[0039] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0040] Figure 1 This is a schematic diagram of the overall structure of an intelligent system paradigm operating system proposed in this invention;

[0041] Figure 2 This is a diagram showing the three core functional domain structures of an intelligent system paradigm operating system and its implementation method proposed in this invention.

[0042] Figure 3 This is a flowchart illustrating the system loop and state update process of an intelligent system paradigm operating system and its implementation method proposed in this invention.

[0043] Figure 4 This diagram illustrates the unified data protocol and trust bridge communication mechanism of an intelligent system paradigm operating system and its implementation method proposed in this invention.

[0044] Figure 5 This is a schematic diagram of the cross-platform operation and expansion interface of the intelligent system paradigm operating system and its implementation method proposed in this invention. Detailed Implementation

[0045] The technical inventions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0046] It should be noted that the terms “front,” “back,” “left,” “right,” “up,” and “down” used in the following description refer to the directions shown in the attached diagram, while the terms “inside” and “outside” refer to the directions toward or away from the geometric center of a specific component, respectively.

[0047] like Figures 1-5As shown, the technical invention adopted in this invention is as follows: an intelligent system paradigm operating system, the system as a whole possesses intelligent system attributes, and the system has a built-in and continuously running system-level intelligent loop consisting of six stages: perception, cognition, decision-making, execution, feedback, and learning, at the operating system layer; the system includes the following three core functional domains:

[0048] The perception-driven domain (L1) serves as the physical and execution foundation layer of the system. It is responsible for multimodal signal acquisition, such as through sensor driving, system call monitoring, hardware and software resource scheduling, issuance and execution of specific task execution instructions, and collection of task execution results and system status data as raw feedback. It can be implemented by the kernel module of the operating system, device driver framework and resource manager.

[0049] The cognitive decision domain (L2) serves as the core intelligent layer of the system. It is responsible for semantic understanding, contextual analysis, logical reasoning, and task planning of the perceptual information received from L1, generating specific decision-making schemes. Simultaneously, it analyzes the feedback data collected by L1 and updates the cognitive model and decision-making strategy through built-in learning algorithms such as reinforcement learning and parameter optimization, achieving self-optimization. It can be implemented as one or a group of intelligent engines within an operating system kernel, such as integrating a lightweight neural network inference engine, a rule engine, and a knowledge graph.

[0050] Interactive Expression Domain (L3): As the interface layer for the interaction between the system and the internal and external environment, it is responsible for receiving user input such as voice, touch, external system commands or environmental signals, and transmitting them into the system after standardization; at the same time, it is responsible for outputting the system's internal decision results and status information to the user or external entities in multimodal forms such as graphical interface, voice, and protocol messages. It can be built based on the operating system's input / output subsystem, graphics services and network communication stack.

[0051] like Figure 4 As shown, the three functional domains are connected and coordinated through two core mechanisms:

[0052] Unified Data Protocol (UDP / UDF): Defines the message format, data structure, authentication fields, and status coding rules for cross-domain communication, ensuring the syntactic consistency of data flow between L1, L2, and L3, including but not limited to the data layer, authentication layer, and trust layer structures;

[0053] Trust Bridge Mechanism (TBM): As a security pillar spanning the three domains, it performs mandatory identity authentication, permission verification, message integrity protection, and encrypted transmission for all cross-domain communications, and maintains a traceable trust chain to ensure that the operation of each link in the intelligent loop is authorized and verifiable.

[0054] All cross-domain and cross-terminal communications are authenticated, encrypted, and verified via a trust bridge to ensure consistent permissions, data trustworthiness, and execution security, thereby achieving system-level trust continuity and multi-node collaborative operation.

[0055] Any mechanism that can achieve message integrity, encrypted communication, caller verification, and state synchronization constitutes an equivalent implementation of UDP / UDF and TBM.

[0056] The system can be deployed and run on different software and hardware platforms, and maintains state synchronization and trust consistency with TBM in multi-terminal and multi-node environments through UDP / UDF.

[0057] The deployment topology includes, but is not limited to, centralized, edge-first, end-to-cloud collaborative, or federated mesh structures. Any operating mode that can maintain the continuity of intelligent cycles and the consistency of feedback constitutes an equivalent implementation of the present invention.

[0058] The perception-driven domain (L1) includes, but is not limited to, signal acquisition, resource scheduling, task execution and feedback acquisition modules, which are responsible for realizing the logical closure of the perception, execution and feedback stages in the system loop, and constitute the physical execution and perception foundation layer of the system.

[0059] The cognitive decision domain (L2) includes, but is not limited to, semantic understanding, logical reasoning, task planning, trust assessment and learning optimization modules. This domain generates decision plans based on feedback from L1 and input from L3, and achieves resource coordination and strategy transfer through UDP / UDF and TBM, forming the core cognitive and learning layer of the system loop.

[0060] The interactive expression domain (L3) includes, but is not limited to, input acquisition, output expression, multimodal interaction and external communication modules, which are used to connect users, environment and external systems to form a closed-loop interactive mechanism of input-execution-feedback-output.

[0061] The system center serves as a unified hub connecting various domains and supports the addition of new scalable system domains (Lx) to accommodate future system-level functional expansions. External functional units (EFU / CU) connect to the system kernel via extension interfaces (LEI), providing the system with open capability access.

[0062] Any external function call mechanism that has security verification, capability registration, and state synchronization constitutes an equivalent implementation of LEI and EFU.

[0063] L1 Perception-Driven Domain: Acquires signals, monitors status, schedules resources, and generates feedback; L2 Cognitive-Decision Domain: Performs semantic understanding, logical reasoning, strategy planning, and learning optimization; L3 Interaction-Expression Domain: Responsible for multimodal interaction, external functional unit invocation, and ecosystem interface management; The three domains work together to form a unified intelligent loop engine for the system, achieving state consistency and adaptive operation capabilities across devices, environments, and tasks.

[0064] like Figure 5 As shown, the system operates and expands in a multi-platform environment: The three domains adopt a modular and abstract design, which can be deployed in a variety of software and hardware environments, and realize system capability expansion, external capability access and ecosystem-level collaboration through hierarchical extension interfaces (LEI) and external functional units (EFUs).

[0065] The system can abstract external functional modules into unified calling units (Capability Units), forming a loop mechanism of "user → system → external functional unit (EFU) → system state write-back", realizing long-term evolution and sustainable operation of the distributed intelligent system.

[0066] This invention proposes an implementation method for an intelligent system paradigm operating system. The operation process of this system is as follows: Figure 1-5 As shown, the interactive expression domain (L3), the perception-driven domain (L1), and the cognitive decision-making domain (L2) form a closed loop according to the collaborative rhythm of "L3→L1→L2→L2→L1→L1→L2→L3". The system-level intelligent loop implemented at the operating system layer includes the following steps:

[0067] Step 1: L3 input stage. The interactive expression domain (L3) receives the raw input signal through input devices such as microphone, camera, keyboard or network interface, performs preprocessing and standardization, verifies it through the Trust Bridge Mechanism (TBM), encapsulates it according to the unified data protocol (UDP / UDF), and sends it to the perception driving domain (L1).

[0068] Step 2: In the L1 perception phase, the perception driver domain (L1) receives standardized data packets from L3 and detects the current resource status of the system, such as CPU utilization, remaining memory, and network bandwidth, through the status monitoring module. It then generates a system perception report based on the input signals. The perception report is encrypted by TBM and transmitted to L2. At the same time, L1 initializes the task execution resource pool to prepare for subsequent task execution.

[0069] Step 3: In the L2 cognitive stage, the semantic understanding module of the cognitive decision domain (L2) analyzes the input signals in the perception report, identifies user intentions or environmental needs such as "adjust the indoor temperature to 26℃", and generates a task semantic model; the logical reasoning module combines the system's historical state data and trust assessment results to determine the resources and external units required for task execution (such as calling the smart air conditioning module).

[0070] Step 4: In the L2 decision-making phase, the L2 task planning module, based on the semantic model and resource assessment, formulates specific execution strategies such as "calling EFU-intelligent air conditioner, allocating 10Mbps bandwidth, and executing temperature adjustment instructions," and generates a decision data packet containing execution instructions and authorization information; the decision data packet is then transmitted to L1 after being signed by TBM.

[0071] Step 5: During the L1 execution phase, the L1 resource scheduling module allocates system resources according to the decision instructions and calls the specified external functional unit (EFU) through the extended interface (LEI) to control the EFU to execute specific tasks. During task execution, L1 collects execution status data in real time (such as air conditioner operating power and temperature change curves).

[0072] Step 6: L1 Feedback Phase. After the task is completed, the L1 feedback acquisition module integrates the execution result data, such as "the indoor temperature has reached 26℃, taking 3 minutes", with the system resource consumption data to generate a feedback report. The feedback report is transmitted to L2 after being standardized by UDP / UDF and encrypted by TBM.

[0073] Step 7: In the L2 learning phase, the L2 learning optimization module analyzes the feedback report. If the task execution meets the target, such as temperature adjustment error ≤ 0.5℃, the historical task database is updated to strengthen the corresponding decision model. If the target is not met, such as adjustment timeout, the decision strategy is optimized through reinforcement learning algorithm, such as "prioritizing the allocation of higher bandwidth when the air conditioner is called next time". The updated model parameters are stored in the system knowledge base.

[0074] Step 8: L3 output stage. L2 transmits the optimization results and execution feedback to L3. The L3 output encoding module converts the information into multimodal forms, such as voice broadcast "The temperature has been adjusted to 26℃" and APP interface display of adjustment records, and feeds back to the user or external system to complete one intelligent cycle.

[0075] like Figure 2 and Figure 3 As shown, all data flows in the above process must be verified by the Trust Bridge Mechanism (TBM) under the premise of conforming to the Unified Data Protocol (UDP / UDF). The system completes a full six-stage intelligent cycle under the cooperation of the three domains: perception → cognition → decision-making → execution → feedback → learning.

[0076] The three domains operate collaboratively at the system level in a distributed manner, wherein:

[0077] L3 is responsible for input and output (loop start and end points);

[0078] L1 handles perception and execution (implemented at the physical layer);

[0079] L2 enables cognition, decision-making, and learning (the core of the intelligence layer).

[0080] Ultimately, the system forms a unified closed-loop mechanism at the operating system layer, realizing an intelligent system structure that integrates perception, cognition, and interaction.

[0081] The system's intelligent loop mechanism and state update path:

[0082] 1. The input signal is acquired by L3 and transmitted to the system;

[0083] 2. L1 performs signal sensing and resource monitoring;

[0084] 3. L2 performs semantic understanding, logical reasoning, and task decision-making based on perceptual information;

[0085] 4. L1 allocates resources to execute tasks and generates feedback data based on the decision results;

[0086] 5. Feedback data is sent back to L2 for learning and strategy optimization;

[0087] 6. The optimization results are output via L3 to respond or trigger external behaviors.

[0088] This process corresponds to a six-stage intelligent cycle at the system level: perception → cognition → decision-making → execution → feedback → learning. This cycle runs continuously at the operating system level, constituting a system-level self-learning and self-optimization mechanism, achieving task continuity, state stability, and intelligent evolution capabilities.

[0089] Any system structure or implementation method that possesses two or more intelligent functions, including but not limited to perception, cognition, decision-making, execution, feedback, and learning, and can form an intelligent logical closed loop at the operating system layer, regardless of changes in naming, architecture, module division, communication protocol, or technical implementation path, belongs to the equivalent technical solution of this invention. The naming, boundaries, or deployment forms of L1–L3 may vary. Any entity that undertakes equivalent perception-driven, cognitive decision-making, and interactive expression functions is considered to be the equivalent mapping domain of this invention.

[0090] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, material, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, material, or apparatus.

[0091] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent system paradigm operating system, characterized in that, The system as a whole possesses the attributes of an intelligent system. Built into the operating system layer, it continuously runs a system-level intelligent loop consisting of six stages: perception, cognition, decision-making, execution, feedback, and learning. The system includes the following three core functional domains: The perception-driven domain (L1) is used for acquiring signals, scheduling resources, and generating feedback. Cognitive decision domain (L2): used for semantic understanding, logical reasoning, task planning, and learning optimization; Interactive expression domain (L3): used for multimodal input / output and external communication; The three core functional domains communicate and synchronize their states through a unified data protocol (UDP / UDF) and a trust bridge mechanism (TBM), enabling cross-domain authentication, trust transfer, and system-level intelligent looping. The three core functional domains can be logical domains, physical instances, or distributed nodes, and can run in same-machine, cross-machine, or multi-instance environments to maintain system cycle and state consistency.

2. The intelligent system paradigm operating system according to claim 1, characterized in that, The perception-driven domain (L1) includes, but is not limited to, signal acquisition, resource scheduling, task execution and feedback acquisition modules, which are responsible for realizing the logical closure of the perception, execution and feedback stages in the system loop, and constitute the physical execution and perception foundation layer of the system.

3. The intelligent system paradigm operating system according to claim 1, characterized in that, The cognitive decision domain (L2) includes, but is not limited to, semantic understanding, logical reasoning, task planning, trust assessment and learning optimization modules. This domain generates decision plans based on feedback from L1 and input from L3, and achieves resource coordination and strategy transfer through UDP / UDF and TBM, forming the core cognitive and learning layer of the system loop.

4. The intelligent system paradigm operating system according to claim 1, characterized in that, The interactive expression domain (L3) includes, but is not limited to, input acquisition, output expression, multimodal interaction and external communication modules, which are used to connect users, environment and external systems to form a closed-loop interactive mechanism of input-execution-feedback-output.

5. The intelligent system paradigm operating system according to claim 1, characterized in that, UDP / UDF defines cross-domain communication formats, authentication processes, and trust transfer logic, including but not limited to the data layer, authentication layer, and trust layer structure. TBM performs identity authentication, access control, trust logging, and authorization transfer, and maintains the continuity of the trust chain and the verifiability of the state throughout the entire system process.

6. The intelligent system paradigm operating system according to claim 1, characterized in that, The system loads external functional units through the Layered Extension Interface (LEI) or the Universal Interface Framework (UIF) to extend the system's functions and task capabilities. These external functional units include, but are not limited to, plugins, services, models, algorithms, or hardware interfaces. After verification by TBM, they can be integrated into the system loop to participate in task execution, status updates, and feedback learning.

7. The intelligent system paradigm operating system according to claim 1, characterized in that, The system can be deployed and run on different software and hardware platforms, and maintains state synchronization and trust consistency with TBM in multi-terminal and multi-node environments through UDP / UDF.

8. The intelligent system paradigm operating system according to claim 1, characterized in that, Any system structure or implementation method that possesses two or more intelligent function cycles (including but not limited to perception, cognition, decision-making, execution, feedback, and learning) and can form an intelligent logic closed loop at the operating system layer, regardless of changes in naming, architecture, module division, communication protocol, or technical implementation path, belongs to the equivalent technical solution of this invention; the naming, boundaries, or deployment forms of L1–L3 may vary, and all those that undertake equivalent perception-driven, cognitive decision-making, and interactive expression functions are considered equivalent mapping domains of this invention.

9. A method for implementing an intelligent system paradigm operating system according to any one of claims 1 to 8, characterized in that, Includes the following steps: (1) Input phase (L3): Receives multimodal input signals from users, devices, or the environment; (2) Sensing stage (L1): Acquire and analyze input signals to generate system sensing state; (3) Cognitive stage (L2): Parsing the input content and generating a task semantic model; (4) Decision-making stage (L2): Formulate and authorize execution strategies based on system status and trust assessment; (5) Execution phase (L1): Schedule resources or call external functional units (EFUs) to execute tasks; (6) Feedback phase (L1): Collect execution results and update system status with TBM via UDP / UDF; (7) Learning phase (L2): Based on feedback optimization models and strategies, self-learning and continuous evolution are achieved; (8) Output phase (L3): The optimization results are fed back to the user or external system in the form of language, interface or multimodal. This creates a system-level intelligent cycle at the system level, which involves perception, cognition, decision-making, execution, feedback, and learning.

10. An apparatus for implementing the method of claim 9, characterized in that, It includes a processor and a memory, wherein the memory stores instructions that, when executed by the processor, cause the processor to perform all the steps of claim 9.

11. A non-transitory computer-readable medium, characterized in that, It stores instructions that, when executed by a processor, cause the processor to perform all the steps of the method of claim 9.

Images