Self-development method, system and storage medium based on morphogenetic field

Abstract

The application provides a self-development method and system based on a morphogenetic field, and a storage medium. The self-development method comprises the following steps: receiving a user intention, converting the user intention into a structured intention message; generating a functional semantic manifold and a mixed manifold space, and constructing a morphogenetic field; determining an implementation mode of a task target; decomposing the task target into a plurality of design tasks, and determining a design task of each morphogenetic engine; for each local state field, a corresponding morphogenetic engine is used to solve it; determining whether a cooperative stable state is reached among the plurality of morphogenetic engines; after the cooperative stable state is reached, collecting achievement data of the morphogenetic engine, and generating a standardized delivery package. Through intention analysis and construction of a morphogenetic field, the hardware and software implementation mode of the task target is objectively determined, the subjective bias of manual disassembly is eliminated, and the delivery process from abstract intention to standardized engineering products is realized.

Description

Technical Field

[0001] This application relates to the field of digital delivery technology, and in particular to a self-development method, system and storage medium based on morphogenesis field. Background Technology

[0002] In the design and implementation of engineering systems, users typically begin by specifying high-level requirements for the target system's functionality, performance, boundary conditions, or usage scenarios. These requirements often appear in the form of natural language, semi-structured text, or constraints. In existing technologies, user intentions... Figure 1 Generally, this requires understanding and breakdown by professionals. On the other hand, in the process of realizing engineering tasks, existing technologies lack a unified mechanism for determining whether task objectives should be achieved through software or hardware design. In interdisciplinary and interprofessional engineering, the same functional objective may involve multiple dimensions such as geometry, materials, control logic, parameter scheduling, and local physical states. If only human experience is relied upon for task division, unreasonable task breakdown and inconsistent implementation paths can easily occur, thereby increasing design costs and reducing collaborative efficiency. Furthermore, the output data of the solution unit usually requires manual processing and format conversion of different types of output data to form deliverable engineering artifacts. This approach not only increases labor costs but also easily leads to the loss of important state information, verification evidence, and contextual data during the solution process at the delivery stage, which is detrimental to subsequent traceability. Therefore, how to achieve a highly efficient, low-cost, and stable automated digital delivery has become an urgent problem to be solved. Summary of the Invention

[0003] In view of this, embodiments of this application provide a self-development method, system, and storage medium based on morphogenesis field, enabling self-development delivery from user intent to standardized engineered artifacts.

[0004] Firstly, a self-development method based on morphogenetic fields is provided for a morphogenetic field-based self-development system. The self-development system includes multiple cooperating morphogenetic engines. The self-development method includes: receiving user intent and converting the user intent into a structured intent message; generating a functional semantic manifold based on the structured intent message; constructing a hybrid manifold space based on the geometric space and the functional semantic manifold; constructing a morphogenetic field on the hybrid manifold space; calculating the first gradient of the morphogenetic field with respect to the spatial coordinate variables of the geometric space, and the second gradient of the morphogenetic field with respect to the functional parameter variables of the functional semantic manifold; and determining the structured intent message based on the first and second gradients. The implementation methods for the task objectives include software design implementation or hardware design implementation. Based on the implementation methods, the task objectives are decomposed into multiple design tasks, and the design tasks of multiple morphology generation engines are determined for each task. The multiple morphology generation engines executing different design tasks correspond to local state fields in the morphology generation field. For each local state field, the corresponding morphology generation engine is used to solve the problem. During the solution process, it is determined whether the multiple morphology generation engines have reached a cooperative stable state. When the multiple morphology generation engines have reached a cooperative stable state, the result data of the morphology generation engines is collected, and a standardized delivery package is generated based on the result data.

[0005] The aforementioned self-development method based on morphogenetic fields converts user intent into structured intent messages and combines this with a morphogenetic field constructed in a hybrid manifold space to drive the self-development architecture to automatically differentiate task objectives. By comparing the first gradient of the geometric space with the second gradient of the functional semantic manifold, the software and hardware implementation methods of the task objectives are objectively determined, reducing reliance on manual task allocation and communication and coordination costs, and effectively lowering the development and design complexity of the engineering system.

[0006] Optionally, the hybrid manifold space is the direct product of a three-dimensional Euclidean space and a functional semantic manifold; wherein, generating the functional semantic manifold includes: mapping the functional requirements in the structured intent message to proxy engineering indicators based on the structured intent message; normalizing the proxy engineering indicators and fusing them to generate the functional semantic manifold.

[0007] Optionally, based on the first gradient and the second gradient, the implementation method of the task objective in the structured intent message is determined, including: when the magnitude of the first gradient is greater than or equal to the magnitude of the second gradient, the implementation method of the task objective is determined to be implemented through hardware design; when the magnitude of the first gradient is less than the magnitude of the second gradient, the implementation method of the task objective is determined to be implemented through software design.

[0008] Optionally, during the process of multiple morphology generation engines solving in their respective local state fields, the morphology generation engine performs at least one of the following operations: broadcasting the state value of the morphology generation field through a first communication channel; receiving structured intent messages through a second communication channel; and sending a negotiation request through the second communication channel.

[0009] Optionally, for each local state field, the corresponding morphology generation engine is used to solve the problem. The method also includes: judging the running state of the morphology generation engine based on a preset evaluation mechanism, and initiating a negotiation request when the running state is abnormal. The preset evaluation mechanism includes: calculating the consistency verification residual between any morphology generation engines on the shared physical interaction interface; and determining the running state as abnormal when the consistency verification residual is greater than or equal to a first preset threshold.

[0010] Optionally, the preset evaluation mechanism also includes: calculating the field entropy change rate of multiple morphology generation engines respectively; calculating the consistency verification residual between multiple morphology generation engines on the shared physical interaction interface respectively; calculating the health score of multiple morphology generation engines based on multiple field entropy change rates and multiple consistency verification residuals respectively; and determining the running state as abnormal when any health score is less than a second preset threshold.

[0011] Optionally, achieving a collaborative stable state includes: the consistency verification residual between any morphology generation engines on the shared physical interaction interface is less than a first preset threshold; the health scores of all morphology generation engines are greater than or equal to a second preset threshold; and there are no unprocessed negotiation requests.

[0012] Optionally, the process involves collecting the output data from the morphology generation engine and generating a standardized delivery package based on the output data. This includes: selecting a standard template according to the type of task objective; inputting the output data into the standard template to generate engineering data corresponding to the task objective; and generating a standardized delivery package based on the engineering data.

[0013] Secondly, a self-developing system based on a morphogenetic field is provided, comprising: a front-end intent proxy module configured to receive user intents and generate structured intent messages; a field construction module configured to generate a functional semantic manifold based on the structured intent messages, construct a hybrid manifold space based on the geometric space and the functional semantic manifold, and construct a morphogenetic field on the hybrid manifold space; and a coordination controller configured to calculate the first gradient of the morphogenetic field with respect to the spatial coordinate variables of the geometric space, and the second gradient of the morphogenetic field with respect to the functional parameter variables of the functional semantic manifold, and determine the implementation method of the task objective in the structured intent message based on the first and second gradients. The implementation methods include: software design implementation or hardware design implementation. Based on the implementation method, the task objective is decomposed into multiple design tasks. Multiple morphology generation engines are configured to execute the design tasks respectively and solve the corresponding local state fields in the morphology generation field. The coordination controller is also configured to determine the design tasks of the multiple morphology generation engines respectively, and to determine whether the multiple morphology generation engines have reached a cooperative stable state during the solving process. The delivery module is configured to collect the result data of the morphology generation engines after the multiple morphology generation engines have reached a cooperative stable state, and generate a standardized delivery package based on the result data.

[0014] Thirdly, a computer-readable storage medium is provided having instructions stored thereon, which, when executed by a processor, implement the self-development method based on a morphogenesis field as described in any of the first aspects. Attached Figure Description

[0015] The following is a brief introduction to the accompanying drawings used in the description of the embodiments of this application: Figure 1 A flowchart of a self-development method based on a morphogenesis field provided in some embodiments of this application is shown; Figure 2 A schematic diagram of a self-developing system based on a morphogenesis field provided in some embodiments of this application is shown. Detailed Implementation

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, examples of implementation methods of this application will be described below with reference to the accompanying drawings. The accompanying drawings described below are merely some embodiments of this application. For those skilled in the art, other drawings and other implementation methods can be obtained based on these drawings without creative effort. Adjustments and improvements made without departing from the concept of this application are all within the protection scope of this application.

[0017] To keep the drawings simple, each figure only schematically shows the parts relevant to the embodiment, and they do not represent the actual structure of the product. In addition, for the sake of clarity and ease of understanding, some figures only schematically show parts of components with the same structure or function, and there may actually be more or fewer components with the same structure or function.

[0018] In this application, unless otherwise expressly specified and limited, ordinal numbers, such as "first," "second," etc., are used only to distinguish and describe related objects, and should not be construed as indicating or implying the relative importance or order between related objects; furthermore, they do not represent the quantity of related objects. "Multiple" includes two or more, and other quantifiers are similar. " / " is used to describe the relationship between related objects, indicating an "or" relationship between them. "And / or" is used to describe the relationship between related objects, including any combination relationship between them, such as "a and / or b" including: "a alone," "b alone," or "a and b." "One or more" or "at least one" of multiple objects refers to any object or any combination of multiple objects, such as "one or more of a1, a2, a3" or "at least one of a1, a2, a3" including: "a1 alone," "a2 alone," "a3 alone," "a1 and a2," "a1 and a3," "a2 and a3," or "a1, a2 and a3."

[0019] In the design, simulation, and implementation of engineering projects, modern products are increasingly characterized by cross-domain complexity, logical intricacy, and shortened development cycles. A single product integrates microelectronic circuits, physical structures, control algorithms, and multi-dimensional physical environment constraints. In traditional workflows, these diverse design requirements are typically discretized, requiring manual intent decomposition and task allocation. However, as system complexity increases, this approach exhibits significant limitations in handling ambiguous user intents, cross-domain physical conflicts, and achieving full lifecycle traceability.

[0020] To address the aforementioned issues, this application proposes a self-development method, system, and storage medium based on morphogenetic fields to achieve automatic differentiation and closed-loop delivery of engineering tasks. In this engineering architecture, the system is defined as a living engineering entity with self-organization, self-healing, and continuous development capabilities. By simulating biological development processes, and using morphogenetic fields as the core mathematical foundation, the user's functional expectations are transformed into an evolutionary dynamic distributed in a multi-dimensional space. The morphogenetic engine, as the basic unit for executing evolution, simulates functional cells within a biological organism, collaboratively solving the local states of the field in a distributed manner to drive the system towards a steady state, ultimately realizing the transformation from user intent to engineered artifacts.

[0021] The following description is in conjunction with the accompanying drawings: Figure 1 A flowchart of a morphogenetic field-based self-development method provided in some embodiments of this application is shown. This method is used in a morphogenetic field-based self-development system, wherein the self-development system includes multiple cooperating morphogenetic engines. The method includes at least the following steps: S110: Receive user intent and convert the user intent into a structured intent message; S120: Based on structured intent messages, generate functional semantic manifolds, construct hybrid manifold spaces based on geometric spaces and functional semantic manifolds, and construct morphogenesis fields on hybrid manifold spaces; S130: Calculate the first gradient of the morphogenetic field with respect to the spatial coordinate variables of the geometric space, and the second gradient of the morphogenetic field with respect to the functional parameter variables of the functional semantic manifold; S140: Based on the first gradient and the second gradient, determine the implementation method of the task objective in the structured intent message, wherein the implementation method includes: software design implementation or hardware design implementation; S150: Based on the implementation method, the task objective is decomposed into multiple design tasks, and the design tasks of multiple morphology generation engines are determined respectively. Among them, the multiple morphology generation engines that execute different design tasks correspond to the local state fields in the morphology generation field. S160: For each local state field, the corresponding morphogenetic engine is used to solve it; S170: During the solution process, determine whether multiple morphological engines have achieved a cooperative stable state; S180: After multiple morphogenetic engines reach a coordinated and stable state, collect the result data of the morphogenetic engines and generate a standardized delivery package based on the result data.

[0022] In the above embodiments, the user's intent is first received and converted into a machine-readable intent message. In actual engineering development, user intents typically manifest as functional expectations, performance constraints, or boundary conditions for a product in a specific usage scenario. These intents often exist in the form of natural language, semi-structured text, or requirement items, exhibiting strong ambiguity. This application can perform deep semantic parsing through a front-end intent proxy, transforming this ambiguous description into computable mathematical objects. For example, unstructured terms can be converted into standardized engineering metrics, mapping textual descriptions to machine-readable sets containing target dimensions and thresholds, thereby providing precise input for subsequent mathematical modeling.

[0023] Subsequently, a functional semantic manifold is generated through structured intent messages. The functional semantic manifold is an abstract mathematical space composed of various functional attributes as coordinate axes, where functional attributes can include heat flux density, voltage ripple, etc. This application constructs a high-dimensional hybrid manifold space by combining physical geometric space with this functional semantic manifold. In this hybrid space, each coordinate point not only represents a location in the physical world but also corresponds to a specific set of functional states. Furthermore, a morphogenesis field can be constructed based on the hybrid manifold space to characterize the potential energy function representing the degree of completion of the user's intent within the hybrid manifold space. In the initial state, the morphogenesis field can be viewed as a developmental blueprint, which indicates how the system should evolve to achieve the user's intent through the level of potential energy.

[0024] After the field model is constructed, the task objectives in the structured intent message can be automatically differentiated by calculating the field gradient. The gradient, in a physical sense, represents the drastic change in field values ​​and reflects the evolutionary pressure the system experiences in different dimensions. Therefore, the first gradient of the morphogenetic field with respect to the spatial coordinate variables of the geometric space, and the second gradient of the morphogenetic field with respect to the functional parameter variables of the functional semantic manifold, are further calculated. Based on the first and second gradients, the implementation method of the task objective in the structured intent message is determined, i.e., whether the task should be attributed to software design implementation or hardware design implementation. If the first gradient of a task objective in the spatial coordinate dimension is significant, it indicates that the achievement of the objective depends on changes in the physical structure or adjustments in the spatial layout, and it can be determined that it is implemented by hardware design. If the second gradient of the task objective in the functional parameter dimension is more significant, it indicates that the objective is more efficient through adjustment of logical parameters or algorithm optimization, and it can be determined that it is implemented by software design.

[0025] After determining the method for achieving the task objective, it can be decomposed into multiple specific design tasks, and design tasks for multiple morphogenetic engines can be determined for each task. At this point, the multiple morphogenetic engines executing different design tasks correspond to local state fields within the morphogenetic field. Each morphogenetic engine is considered an autonomous, self-solving engineering entity responsible for maintaining and evolving a specific subdomain within the morphogenetic field, thereby achieving the task objective by simulating the division of labor and cooperation among cells during the development of biological tissues. For each local state field, the corresponding morphogenetic engine is used to solve the problem. The morphogenetic engine continuously iterates to find the optimal solution that satisfies the local boundary conditions and the intended potential energy.

[0026] During the solution process, a unified interactive architecture continuously assesses whether multiple morphogenesis engines have reached a cooperative stable state. This cooperative stable state is the core criterion for measuring whether development is complete. It requires that each morphogenesis engine not only achieves convergence internally but also that the parameters are coupled consistently at the physical or logical interfaces between the morphogenesis engines. Once multiple engines have reached a cooperative stable state, the resulting data can be collected and automatically packaged according to industry standard templates to generate a delivery package containing standardized engineering artifacts.

[0027] The self-development method based on morphogenesis fields provided in this application performs quantitative field gradient calculations within a hybrid manifold space based on structured intent messages, changing the traditional model of task division relying on manual intervention in engineering design. By introducing the first gradient of spatial variables and the second gradient of functional parameter variables into a unified hybrid manifold space, the method achieves automated differentiation of task objectives and objective determination of software and hardware implementation methods at the architecture level. This effectively solves technical problems such as semantic distortion, interdisciplinary design conflicts, and lengthy R&D cycles caused by manual intent decomposition. It lays the technical foundation for subsequent distributed collaborative solving of multiple morphogenesis engines, boundary consistency verification, and automatic encapsulation of standardized engineering results, thereby realizing self-development from intent to standardized engineering artifacts.

[0028] In some embodiments of this application, the hybrid manifold space is the direct product of a three-dimensional Euclidean space and a functional semantic manifold; wherein, generating the functional semantic manifold includes: mapping the functional requirements in the structured intent message to proxy engineering indicators based on the structured intent message; normalizing the proxy engineering indicators and fusing them to generate the functional semantic manifold.

[0029] Based on the above embodiments, the hybrid manifold space is defined as the direct product of a three-dimensional Euclidean space and a functional semantic manifold. When generating the functional semantic manifold, the structured intent message is first received and parsed, mapping each functional requirement to specific proxy engineering metrics. For example, heat dissipation requirements can be mapped to surface heat flux density thresholds using lookup tables or pre-trained natural language models, and signal integrity requirements can be mapped to multiple metrics such as voltage and impedance. Since the physical dimensions and value ranges of different proxy engineering metrics vary greatly, they need to be normalized. For example, for metrics with clear upper and lower limits, a linear extremum normalization method is used to map them to the interval between zero and one; for metrics following a specific distribution, a standardization method is used to eliminate mean drift. Then, according to pre-set weights, the normalized metrics are fused to generate the functional semantic manifold. Furthermore, a mathematical direct product operation is performed between the three-dimensional geometric coordinate system and the functional semantic manifold to generate the hybrid manifold space, unifying the performance metrics of the geometric space and multiphysics fields under the same framework, providing a computable underlying numerical foundation for subsequently solving the gradient of the morphogenetic field.

[0030] In some embodiments of this application, the implementation method of the task objective in the structured intent message is determined based on the first gradient and the second gradient, including: when the magnitude of the first gradient is greater than or equal to the magnitude of the second gradient, the implementation method of the task objective is determined to be implemented through hardware design; when the magnitude of the first gradient is less than the magnitude of the second gradient, the implementation method of the task objective is determined to be implemented through software design.

[0031] Based on the above embodiments, the first and second gradients of the current morphological field are calculated in real time, and their absolute magnitudes are obtained. The method of achieving the task objective is determined by comparing the magnitudes of the gradients. For example, the magnitudes of the first and second gradients can be compared. When the magnitude of the first gradient is greater than or equal to the magnitude of the second gradient, it indicates that achieving the current task objective relies more on changes in physical geometry, thus determining that the current task objective is achieved through hardware design. When the magnitude of the first gradient is less than the magnitude of the second gradient, it indicates that adjusting the logic and control parameters is more effective, thus determining that the current task objective is achieved through software design. For example, in temperature control design, if the magnitude of the first gradient is greater than the magnitude of the second gradient, the task is automatically assigned to hardware design, and the cooling purpose is achieved by increasing the area of ​​the heat sink. By setting an automatic task allocation mechanism, the uncertainty of manual task allocation is eliminated, and the response speed is faster.

[0032] In some embodiments of this application, during the process of multiple morphology generation engines solving in their respective local state fields, the morphology generation engine performs at least one of the following operations: broadcasting the state value of the morphology generation field through a first communication channel; receiving a structured intent message through a second communication channel; and sending a negotiation request through the second communication channel.

[0033] Based on the above embodiments, multiple morphology generation engines initiate parallel solution processes after being assigned to a local state field. During the solution process, the morphology generation engine broadcasts the state values ​​of the morphology generation field at high frequency through a first communication channel and receives structured intent messages and sends negotiation requests through a second communication channel. Furthermore, a unified interaction architecture including multiple communication channels can be set, the core of which consists of three independent and strictly isolated buses. The first bus, i.e., the first communication channel, is dedicated to carrying continuous state signals of the morphology generation field; the second bus, i.e., the second communication channel, is dedicated to transmitting structured intent messages, constraint rules, and exception negotiation requests; and the third bus, as the third communication channel, is independent of the other two service buses and is dedicated to providing runtime state observation and external security intervention capabilities. Furthermore, underlying constraints are set for this unified interaction architecture, for example, stipulating that any two of the above communication channels cannot share message formats, and prohibiting any morphology generation engine from creating or using unauthorized communication links that bypass this unified interaction architecture. In the operational flow, the morphology generation engine calls the first communication channel to broadcast the latest morphology generation field state values ​​to the cooperative network, while continuously listening to and receiving structured intent messages through the second communication channel. Once a local conflict is detected, the morphology engine can immediately send a negotiation request outward through a second communication channel. By physically isolating high-volume data synchronization, logical control, and security intervention, it ensures that high-frequency synchronization data does not block the transmission of intent messages and negotiation requests.

[0034] In some embodiments of this application, for each local state field, a corresponding morphology generation engine is used to solve the problem. The method further includes: judging the running state of the morphology generation engine based on a preset evaluation mechanism; and initiating a negotiation request when the running state is an abnormal state. The preset evaluation mechanism includes: calculating the consistency verification residual between any morphology generation engines on the shared physical interaction interface; and determining the running state as an abnormal state when the consistency verification residual is greater than or equal to a first preset threshold.

[0035] Based on the above embodiments, multiple morphology generation engines continuously calculate in their respective local state fields and continuously synchronize their current operating state data in the second communication channel. These engines determine their own operating state in real time based on a preset evaluation mechanism. When their operating state is abnormal, they can send a negotiation request to the coordination control unit. The preset evaluation mechanism can be based on the operating state data of the morphology generation engines during operation, such as real-time calculation of the consistency verification residual between any two morphology generation engines on the shared physical interaction interface. The core of consistency verification is to quantify the difference in the state fields of the two morphology generation engines on the shared real interaction interface, which can be measured using the Euclidean norm (L2 norm) of the difference in the local multi-physics fused state vectors of the two engines on the shared physical interaction interface. The consistency verification residual can be calculated using the L2 norm of the difference in the fused state vectors of the morphology generation engines on the shared physical interaction interface. The fused state vector can be composed of multi-physics components such as electric field strength, temperature, and stress tensor, and is discretely represented at multiple sampling points on the shared physical interaction interface. The consistency verification residual is determined by summing the squares of the differences between the state components at the corresponding sampling points and taking the square root. The consistency verification residual is used to quantify the difference in the state field of the two morphogenesis engines on the shared physical interaction interface and to reflect the degree of energy mismatch on the interface, thereby providing a basis for the determination of the first type of anomaly.

[0036] For example, suppose two morphogenesis engines (denoted as ME_A and ME_B) that have direct physical interaction provide local multiphysics fusion state vectors on their common interface Γ: Where E is the electric field intensity vector and T is the temperature scalar. These are the stress tensor components. The above physical quantities can be found at the interface. The above is discretized into N sampling points (such as finite element nodes), therefore It can be represented as a multidimensional state vector defined on N sampling points, that is... ,in The state dimension for each point.

[0037] Based on this, the consistency verification residual can be defined as the L2 norm of the difference between two local multi-physics fusion state vectors, referring to Formula 1. Formula 1 Furthermore, Formula 1 above can be written as Formula 2. Formula 2 After calculating the consistency check residual, it is compared with a first preset threshold. The first preset threshold represents the upper limit of the state field difference between the two morphology generation engines on the shared physical interaction interface; its specific value can be set. When the consistency check residual exceeds the first preset threshold, the operating state of the morphology generation engine is determined to be abnormal. A negotiation request for the shared physical interaction interface can be initiated through the second communication channel, submitting this abnormal situation to the coordination and control unit, which then initiates subsequent negotiation processing.

[0038] In some embodiments of this application, the preset evaluation mechanism further includes: calculating the field entropy change rate of multiple morphology generation engines respectively; calculating the consistency verification residual between multiple morphology generation engines on the shared physical interaction interface respectively; calculating the health score of multiple morphology generation engines based on multiple field entropy change rates and multiple consistency verification residuals respectively; and determining the running state as an abnormal state when any health score is less than a second preset threshold.

[0039] Based on the above embodiments, the morphology generation engine can calculate its own field entropy change rate in real time during operation, and combine it with the consistency verification residuals with other morphology generation engines on the shared physical interaction interface to calculate a health score, which serves as the judgment criterion for a preset evaluation mechanism. Here, field entropy is used to quantify the degree of disorder or uncertainty in the local state field of the morphology generation engine. For a discretized fusion state field... The field entropy H of a distribution across N grid points can be defined as the Shannon entropy, as shown in Equation 3. Formula 3 in, Representing the normalized energy density, using the electric field as an example, refer to Formula 4. Formula 4 The rate of change of field entropy can be obtained by time difference, referring to Formula 5. Formula 5 in, This represents the sampling period. The rate of change of field entropy has a clear physical meaning. If the value continues to increase, it indicates that the solution process inside the engine is diverging, and the evolutionary vitality of the system is decreasing.

[0040] Simultaneously, consistency verification residuals between multiple morphogenetic engines on the shared physical interaction interface can be extracted synchronously. These consistency verification residuals can be measured using the Euclidean norm of the difference between local multi-physics fusion state vectors. A larger residual value indicates less conservation of energy or information exchange between adjacent morphogenetic engines on the interaction interface. By calculating the health scores of multiple morphogenetic engines based on multiple field entropy change rates and multiple consistency verification residuals, a comprehensive index measuring the stability of the current operating state of the morphogenetic engines is obtained. Its value range can be normalized and restricted to between zero and one. This application can use an exponential decay model to calculate the health score, as shown in Formula 6. Formula 6 HS represents the health score. Represents the rate of change of field entropy. Represents the consistency check residual; and This represents the sensitivity coefficient, which can be set and adjusted according to actual conditions. This is the tolerance threshold for the rate of change of field entropy, i.e., the first tolerance threshold; This is the tolerance threshold for the consistency verification residual, also known as the second tolerance threshold. The exponential decay model ensures that the health of the morphogenesis engine will only have a substantial negative impact when the rate of change of field entropy or the consistency verification residual exceeds the set tolerance threshold.

[0041] When the health score of any morphogenesis engine is found to be lower than a second preset threshold, the engine's operating state is determined to be abnormal, and a negotiation request is sent through the second communication channel. The morphogenesis engine can include its current health score, field entropy change rate, and consistency verification residual as decision evidence in the negotiation request. Upon receiving this evidence, the coordination and control unit can perform arbitration based on the engine's historical trust weights and trigger subsequent task takeover and updates.

[0042] In some embodiments of this application, achieving a cooperative stable state includes: the consistency verification residual between any morphology generation engines on the shared physical interaction interface is less than a first preset threshold; the health scores of all morphology generation engines are greater than or equal to a second preset threshold; and there are no unprocessed negotiation requests.

[0043] Based on the above embodiments, determining whether a cooperative stable state has been achieved can be set to require the simultaneous satisfaction of multiple dimensions of indicators. For example, the consistency verification residual between any morphology generation engines on the shared physical interaction interface is checked, and it is determined whether the consistency verification residual is strictly less than a first preset threshold, thereby confirming the boundary state of each local state field. Next, the internal health scores of all morphology generation engines are checked, and it is determined whether the health scores of all engines are greater than or equal to a second preset threshold, thereby confirming that all morphology generation engines are in a fully converged state. Then, the communication network is comprehensively checked to confirm whether there are any unprocessed negotiation requests. Only when these three conditions are met simultaneously can it be finally determined that all morphology generation engines have achieved a cooperative stable state.

[0044] In some embodiments of this application, the process of collecting output data from a morphogenesis engine and generating a standardized delivery package based on the output data includes: selecting a standard template according to the type of task objective; inputting the output data into the standard template to generate engineering data corresponding to the task objective; and generating a standardized delivery package based on the engineering data.

[0045] Based on the above embodiments, when all morphogenesis engines reach a steady state and trigger delivery instructions, an automated process for exporting and packaging the output data can be initiated. First, the final output data is collected from each morphogenesis engine. This output data covers information from multiple physical domains, such as electromagnetic property data, thermodynamic property data, structural mechanical property data, and control logic data. Next, the type of the current task objective is identified, and the corresponding industry standard template is automatically retrieved from the system knowledge base, such as a Gerber template for printed circuit boards or a STEP template for mechanical structures. Then, the collected multi-domain output data is format-mapped and input into the standard template to generate engineering data that can be directly read by the factory. Simultaneously, the underlying state information is extracted, including tensor data of each local state field, morphogenesis engine operation configuration parameters, and trust weights accumulated in the collaborative network. Finally, the above engineering data and underlying state information are jointly packaged to generate a complete and standardized delivery package.

[0046] Furthermore, when packaging the deliverables, an immutable verification evidence chain can be automatically generated and embedded. This verification evidence chain can serve as digital proof of the engineering design, recording in detail the evolutionary trajectory from the input intent to the achievement of a steady state. For example, the verification evidence chain can include the encrypted hash value of the original user intent to ensure that the source of the requirement injection has not been illegally tampered with; or it can include the first and second gradients recorded in the cooperative stable state to explain the objective decision-making basis of the system when differentiating software and hardware tasks; or it can include quantitative indicators such as the consistency verification residuals and health scores finally achieved by each morphogenesis engine to prove that the delivered product meets engineering standards in terms of physical boundaries and internal convergence. By chaining and cryptographically signing the above data according to cryptographic time sequence, the verification evidence chain can be constructed and added to the standardized delivery package. Moreover, the delivery model based on the verification evidence chain facilitates functional safety audits by maintenance personnel in the later stages, providing data credentials for the determination of responsibility and quality certification of the engineering system.

[0047] Figure 2 The diagram illustrates a self-developing system based on a morphogenetic field, as provided in some embodiments of this application. The system includes: a front-end intent proxy module configured to receive user intents and generate structured intent messages; a field construction module configured to generate a functional semantic manifold based on the structured intent messages, construct a hybrid manifold space based on the geometric space and the functional semantic manifold, and construct a morphogenetic field on the hybrid manifold space; and a coordination controller configured to calculate a first gradient of the morphogenetic field with respect to the spatial coordinate variables of the geometric space, and a second gradient of the morphogenetic field with respect to the functional parameter variables of the functional semantic manifold, and to determine the task in the structured intent message based on the first and second gradients. The goal is achieved through either software or hardware design. Based on the implementation method, the task goal is decomposed into multiple design tasks. Multiple morphology generation engines are configured to execute the design tasks respectively and solve the corresponding local state fields in the morphology generation field. The coordination controller is also configured to determine the design tasks of the multiple morphology generation engines and, during the solving process of the multiple morphology generation engines, determine whether the multiple morphology generation engines have reached a cooperative stable state. The delivery module is configured to collect the result data of the morphology generation engines after the multiple morphology generation engines have reached a cooperative stable state and generate a standardized delivery package based on the result data.

[0048] Based on the same technical concept, this application also provides a computer-readable storage medium storing instructions thereon, which, when executed by a processor, implement the self-development method based on morphogenesis field as described in any of the above embodiments.

[0049] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not described in detail or in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Furthermore, the above embodiments can be freely combined as needed.

Claims

1. A self-development method based on morphogenetic fields, characterized in that, For a morphogenetic field-based self-development system, the self-development system comprising multiple synergistic morphogenetic engines, the self-development method comprising: Receive user intent and convert the user intent into a structured intent message; Based on the structured intent message, a functional semantic manifold is generated, a hybrid manifold space is constructed based on the geometric space and the functional semantic manifold, and a morphogenesis field is constructed on the hybrid manifold space; Calculate the first gradient of the morphogenetic field with respect to the spatial coordinate variables of the geometric space, and the second gradient of the morphogenetic field with respect to the functional parameter variables of the functional semantic manifold; Based on the first gradient and the second gradient, the implementation method of the task objective in the structured intent message is determined, wherein the implementation method includes: software design implementation or hardware design implementation; Based on the aforementioned implementation method, the task objective is decomposed into multiple design tasks, and the design tasks of multiple morphology generation engines are determined respectively, wherein the multiple morphology generation engines executing different design tasks correspond to the local state fields in the morphology generation field. For each of the local state fields, the corresponding morphology generation engine is used to solve for them. During the solution process, it is determined whether the multiple morphology generation engines have reached a cooperative stable state; Once multiple morphology generation engines reach a coordinated and stable state, the output data of the morphology generation engines is collected, and a standardized delivery package is generated based on the output data.

2. The self-development method based on morphogenetic field according to claim 1, characterized in that, The hybrid manifold space is the direct product of a three-dimensional Euclidean space and a functional semantic manifold; wherein, generating the functional semantic manifold includes: Based on the structured intent message, the functional requirements in the structured intent message are mapped to proxy engineering indicators; The proxy engineering metrics are normalized and then fused to generate the functional semantic manifold.

3. The self-development method based on morphogenetic field according to claim 1, characterized in that, The step of determining the implementation method of the task objective in the structured intent message based on the first gradient and the second gradient includes: When the magnitude of the first gradient is greater than or equal to the magnitude of the second gradient, the task objective is determined to be achieved through the hardware design. When the magnitude of the first gradient is less than the magnitude of the second gradient, the task objective is determined to be achieved through the software design.

4. The self-development method based on morphogenetic field according to any one of claims 1 to 3, characterized in that, Also includes: During the process of multiple morphology generation engines solving in their respective local state fields, each morphology generation engine performs at least one of the following operations: The state value of the morphogenesis field is broadcast through the first communication channel; The structured intent message is received through the second communication channel; A negotiation request is sent through a second communication channel.

5. The self-development method based on morphogenetic field according to claim 4, characterized in that, The step of solving for each of the local state fields using the corresponding morphology generation engine also includes: Based on a preset evaluation mechanism, the operating status of the morphology generation engine is determined. When the operating status is abnormal, a negotiation request is initiated. The preset evaluation mechanism includes: calculating the consistency verification residual between any of the morphology generation engines on the shared physical interaction interface; when the consistency verification residual is greater than or equal to a first preset threshold, determining the running state as the abnormal state.

6. The self-development method based on morphogenetic field according to claim 5, characterized in that, The preset evaluation mechanism also includes: Calculate the field entropy change rate of each of the aforementioned morphology generation engines; Calculate the consistency verification residuals among the multiple morphology generation engines on the shared physical interaction interface; Based on the multiple field entropy change rates and the multiple consistency verification residuals, the health scores of the multiple morphogenesis engines are calculated respectively. When any of the health scores is less than the second preset threshold, the operating state is determined to be the abnormal state.

7. The self-development method based on morphogenetic field according to claim 6, characterized in that, Achieving the aforementioned cooperative stable state includes: The consistency verification residual between any of the morphology generation engines on the shared physical interaction interface is less than the first preset threshold. The health scores of all the aforementioned morphology generation engines are greater than or equal to the second preset threshold; There are no outstanding negotiation requests.

8. The self-development method based on morphogenetic field according to claim 1, characterized in that, The process of collecting the morphogenesis engine's output data and generating a standardized delivery package based on the output data includes: Select a standard template based on the type of the task objective; Input the output data into the standard template to generate the engineering data corresponding to the task objective; Based on the engineering data, the standardized delivery package is generated.

9. A self-developing system based on a morphogenetic field, characterized in that, include: The front-end intent proxy module is configured to receive user intents and generate structured intent messages; The field construction module is configured to generate a functional semantic manifold based on the structured intent message, construct a hybrid manifold space based on the geometric space and the functional semantic manifold, and construct a morphogenesis field on the hybrid manifold space; A coordination controller is configured to calculate a first gradient of the morphogenesis field with respect to the spatial coordinate variables of the geometric space, and a second gradient of the morphogenesis field with respect to the functional parameter variables of the functional semantic manifold. Based on the first gradient and the second gradient, the controller determines the implementation method of the task objective in the structured intent message, wherein the implementation method includes: software design implementation or hardware design implementation. Based on the implementation method, the task objective is decomposed into multiple design tasks. Multiple morphology generation engines are configured to execute the design task respectively and solve the corresponding local state field in the morphology generation field; The coordination controller is also configured to determine the design tasks of the plurality of morphology generation engines, and, during the solving process of the plurality of morphology generation engines, determine whether the plurality of morphology generation engines have reached a cooperative stable state. The delivery module is configured to collect the output data of the morphology generation engines after the multiple morphology generation engines reach a coordinated and stable state, and generate a standardized delivery package based on the output data.

10. A computer-readable storage medium, characterized in that, It stores instructions that, when executed by a processor, implement the self-development method based on a morphogenesis field as described in any one of claims 1 to 8.

Images