A subtask automatic arrangement method for survey report generation task

Abstract

This invention belongs to the field of exploration report generation technology and provides an automatic subtask arrangement method for exploration report generation. To address the low efficiency and large deviations in existing exploration report generation methods, this invention, based on the principles of multi-step decomposition, single-step generation, multi-candidate selection, real-time verification, and error isolation, achieves intelligent arrangement and stable execution of the entire process from "data input - stratigraphic division - parameter statistics - drawing generation - text writing - report synthesis." It effectively solves industry pain points in existing technologies such as "unreliable large model generation, cross-step error propagation, and difficulty in implementing standard provisions." While maintaining the existing professional software system, it significantly improves the automation level and engineering reliability of exploration report generation, demonstrating significant technological breakthrough value and promising application prospects.

Description

Technical Field

[0001] This invention belongs to the field of survey report generation technology, specifically relating to an automatic subtask arrangement method for survey report generation tasks. Background Technology

[0002] The current production process for geotechnical engineering investigation reports is characterized by a long chain, multiple stages, and high dependence, including field logging, laboratory testing, parameter statistics, comprehensive stratigraphic judgment, drawing generation, text writing for each chapter, and proofreading and compilation. Although mature software exists for each stage, the lack of a unified automated compilation mechanism among these tools leads to the following problems:

[0003] 1. The processing chain is long, requiring continuous manual import / export of data, which is inefficient;

[0004] 2. The various links are weakly coupled, making it impossible to form a "one-click" overall process;

[0005] 3. The report content has complex logic (such as stratigraphic division, bearing capacity inference, and consistency requirements of mandatory provisions), making it difficult to generate a stable large model in one go;

[0006] 4. The content depends on a combination of multiple small tasks. For example, a cross-section requires multiple steps such as stratigraphic identification, thickness calculation, coordinate extraction, and multi-source data fusion. The results then affect subsequent text descriptions, appendix entries, etc.

[0007] If traditional large models directly generate complete chapters or entire reports, the following problems are likely to occur: errors propagate across long texts; inconsistencies in stratigraphic logic; small errors accumulate into large deviations; failure to meet the mandatory requirements of regulatory provisions; and inability to automatically link figures, tables, and text. Summary of the Invention

[0008] To address the technical problems of low efficiency and large deviations in existing survey report generation, this invention provides an automatic subtask arrangement method for survey report generation, which achieves data input, report output, and unattended operation throughout the entire process while retaining existing professional algorithms and mapping kernels.

[0009] To solve the technical problem, the technical solution adopted by this invention is as follows:

[0010] An automatic subtask orchestration method for survey report generation tasks includes:

[0011] (1) Access multi-source heterogeneous data and perform standardization processing. The standardization processing converts multi-source heterogeneous data into a unified structured JSON format, thereby obtaining standardized data.

[0012] (2) The task orchestration engine receives standardized data and starts the task parsing process. Based on the number of strata in the standardized data, the report template selected by the user, and the complexity matrix, the task orchestration engine plans an execution sequence containing 20-80 atomic steps.

[0013] In some embodiments, the difference in the number of atomic steps is based on the following dimensions:

[0014] (2.1) Number of stratigraphic layers: If the site has a simple stratigraphic structure (e.g., only 3 layers), the profile generation only requires about 3-5 atomic steps of "reading data - calculating coordinates - receiving the connection". If the stratigraphic structure is complex (e.g., 15 layers, with lenses, faults or pinch-outs), then steps such as "lens closure processing", "fault line interpolation", and "stratigraphic pinch-out logic judgment" need to be added, which leads to a significant increase in the number of atomic steps.

[0015] (2.2) Report template type: The preliminary exploration report only requires basic logging and statistics (about 20-40 atomic steps), while the detailed exploration report needs to include advanced calculation steps such as liquefaction judgment, corrosivity evaluation, and pile foundation parameter calculation (about 60-80 atomic steps).

[0016] (2.3) Complexity matrix: The complexity matrix is ​​closely related to the number of attached plots. For each additional cross-sectional plot or bar chart, the task parser will automatically add a corresponding sub-sequence of "data extraction - coordinate transformation - plotting instruction generation".

[0017] In some embodiments, during the execution of each atomic step, the state manager manages the dependencies of each atomic step and tracks the execution status in real time; if an atomic step fails, the rollback controller is used to roll back to the nearest stable checkpoint.

[0018] (3) Single-step generation loop: The large language model performs closed-loop processing for each atomic step.

[0019] In some embodiments, the large language model performs closed-loop processing on each atomic step, specifically including:

[0020] (3.1) The big oracle model constructs a context based on the execution structure of the preceding steps and uses the generation module in the big language model to perform parallel sampling to generate 3-8 candidate JSON instructions.

[0021] (3.2) The generated instructions are immediately passed through the exception filter (red flag mechanism), which removes any candidates with format errors or values ​​that exceed the specification boundaries.

[0022] (3.3) For the remaining compliant candidate actions (i.e. JSON instructions), the robust voting device executes the strategy of "stopping when the first K votes are reached" by statistically analyzing semantic redundancy, and quickly selects the optimal action (i.e. the optimal JSON instruction) with the highest consensus.

[0023] (4) The optimal action is verified by a small model, and the verification is completed from at least two dimensions: compliance with engineering specifications and consistency with geological logic. If the verification is successful, the instruction is issued and executed; if it fails, a rollback and retry are triggered.

[0024] In the specific implementation process, the optimal action is verified by a small model, which specifically includes:

[0025] Schema validation: Use Schema Validator to check the integrity of the JSON structure, and use small models to determine whether the data fields conform to the semantic definition of the current step (i.e., the current atomic step).

[0026] Standard compliance verification: Based on the Retrieval Enhanced Generation (RAG) approach, the values ​​in the optimal action (such as "cohesion = 5 kPa") are compared and verified with the provisions of the built-in standard database (such as the "Code for Investigation of Geotechnical Engineering" GB50021).

[0027] Logical consistency verification: Based on natural language reasoning (NU), it checks whether the contextual logic is contradictory. For example, the input pair is: Premise: "The previous layer is the 3rd layer of silty clay", Hypothesis: "This layer is the 2nd layer of silt".

[0028] (5) Verified instructions drive atomic executors to operate. If execution is successful, the global state pool is updated; if execution fails or verification fails, the corresponding rollback strategy is immediately initiated.

[0029] (6) Result Synthesis Output: After all atomic steps are completed, the output synthesis model collects all execution results, arranges the content and optimizes the format according to the exploration report annotation template, and generates the final exploration report.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] The automatic subtask orchestration method for survey report generation in this invention uses a multi-step decomposition principle (that is, decomposing the complex survey report generation into multiple atomic steps). Each step has clear input, processing and output definitions, and the steps are connected through data dependencies.

[0032] Single-step generation principle: Large models generate only one atomic step's action description at a time, avoiding the error propagation problem in traditional chained generation. Each action uses a structured description to ensure parsing and execution. This ensures that the single-step output is highly consistent with engineering specifications.

[0033] The principle of multi-candidate selection: Multiple candidate actions are obtained through multiple samplings. After anomaly filtering, a voting mechanism based on multiple indicators is used to select the optimal solution, which significantly improves the reliability of single-step decision-making.

[0034] Real-time verification principle: A lightweight discrimination model is introduced, and RAG technology is used to search the survey specification database in real time to conduct engineering logic and compliance reviews on candidate actions.

[0035] Error isolation principle: By using atomic executors and rollback mechanisms, local execution exceptions can be captured in time and targeted retries can be triggered, ensuring the robustness of the global task chain.

[0036] In summary, based on the multi-step decomposition principle, single-step generation principle, multi-candidate selection principle, real-time verification principle, and error isolation principle of this invention, this invention has the following specific advantages:

[0037] 1. Significantly improved reliability: Error propagation is effectively suppressed, and the overall logical consistency of the survey report is greatly improved.

[0038] 2. Improved accuracy: Single-step output maintains a high degree of consistency with engineering specifications and experience-based structures.

[0039] 3. Significantly enhanced consistency: Graphs, tables, and queries are generated from the same source, and data with the same name is automatically aligned.

[0040] 4. Overall efficiency leap: The entire process time is significantly shortened compared to the traditional chain method, and manual operation is significantly reduced.

[0041] 5. Stability is fully guaranteed: the output fluctuation is small after multiple runs, and the system can automatically roll back and continue running in case of failure. Attached Figure Description

[0042] Figure 1 This is a flowchart of the present invention;

[0043] Figure 2 This is a flowchart illustrating the single-step (i.e., single atomic step) inner loop sampling, filtering, and rollback process of the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. The described embodiments are some, but not all, of the embodiments of the present invention. The following embodiments are only for specific illustration of the implementation methods of the present invention and do not limit the scope of protection of the present invention.

[0045] Combined with appendix Figure 1 and attached Figure 2 The present invention provides an automatic subtask arrangement method for survey report generation tasks, comprising:

[0046] (1) Access multi-source heterogeneous data and perform standardization processing. Standardization processing involves converting multi-source heterogeneous data into a unified structured JSON format to obtain standardized data. Then, data quality detection is completed and a data quality report is generated, laying the foundation for subsequent processing.

[0047] In the specific implementation process, a data access module is used to access multi-source heterogeneous data. This multi-source heterogeneous data, based on its different sources, includes field logging data and indoor experimental data. Field logging data refers to the original records directly obtained by engineering geological survey personnel at the construction site (field) through drilling, pitting, and other methods. It mainly includes borehole numbers, borehole elevations, drilling depths, descriptions of the soil and rock mass (color, state, humidity, inclusions, etc.), and groundwater level observation records.

[0048] Based on the types of multi-source heterogeneous data, including tabular data, WORD documents, drawings, and unstructured descriptions (such as geotechnical descriptions), the data access module needs to convert different types of data into a unified structured JSON format.

[0049] For tabular data (such as field cataloging and indoor experiments), the system calls a data processing library (such as Python's Pandas or OpenPyXL) to read Excel / CSV source files. Through a pre-defined "field mapping table," inconsistent headers with different units (such as "moisture content" and "moisture percentage (%)") are uniformly mapped to standard system key values ​​(such as water_content); that is, the data processing library converts the tabular data into a unified structured JSON format.

[0050] For unstructured descriptions (such as descriptions of soil and rock masses), regular expressions or NLP entity extraction techniques are used to automatically decompose them into JSON objects of {color: "gray", state: "soft plastic", soil_type: "silty clay"}; that is, using regular expressions or NLP entity extraction techniques to convert unstructured descriptions into a unified structured JSON format.

[0051] In practice, regular expressions convert unstructured descriptions (such as descriptions of soil and rock masses) into a unified structured JSON format, specifically including:

[0052] (1.1.1) Analyze the text structure: Identify the fixed patterns in the data to be extracted (such as key-value pair delimiters, repeating blocks, special markers).

[0053] (1.1.2) Design target JSON structure: Determine the output format (object, nested array, etc.) as needed.

[0054] (1.1.3) Write regular expressions: Use regular expressions to match and capture key data groups.

[0055] (1.1.4) Extraction and transformation: Use hot modules to perform matching and build a dictionary list.

[0056] (1.1.5) Handling complex situations, including: non-fixed order: improve readability by using named grouping;

[0057] Multi-line blocks: Enable the re.DOTALL flag to match newlines;

[0058] Special escaping: Use re.escape() to process text containing regular expression metacharacters.

[0059] (1.1.6) Verification and Output:

[0060] Validate JSON data: Use json.loads(json_output) to test if it can be parsed.

[0061] Improve output quality: use json.dumps(..., indent=2) or customize indentation.

[0062] Regular expressions are not suitable for parsing nested structures (such as HTML / XML); in such cases, a dedicated parser is more reliable. When handling Unicode characters such as Chinese characters, add the `re.UNICODE` flag.

[0063] By following the steps above, text data matched by regular expressions can be systematically converted into structured JSON.

[0064] For drawings (topographic maps / plan maps): The system integrates a DXF parsing library (such as ezdxf or Teigha), directly reading the underlying data entities of DWG / DXF files without relying on a graphical interface. It extracts the circular entity coordinates (X, Y) from key layers (such as the "drill hole location layer") as drill hole coordinates, extracts text annotation entities as drill hole numbers, and finally converts them into geometric data in GeoJSON format; that is, it uses the DXF parsing library to convert the drawings into a unified structured JSON format.

[0065] In the specific implementation process, the drawings (CAD / BIM) will be converted into specific components, including:

[0066] (1.2.1) Requirements analysis and mapping table design, including:

[0067] Identify the input source: Determine the source of the drawings (such as CAD, PDF, scanned drawings) and their data characteristics;

[0068] Define target fields: Based on business requirements, define the fields that need to be included in the final JSON (e.g., "id", "name", "type", "coordinates", "area", "layer").

[0069] Create a mapping table: Establish a field mapping rule for each input source (usually a CSV or JSON configuration file).

[0070] (1.2.2) Raw data extraction and parsing:

[0071] Geometric information: vertex coordinates, line segments, polygon boundaries, etc.;

[0072] Attribute information: layer name, color, line type, associated text annotations, etc.

[0073] (1.2.3) Field alignment and data cleaning:

[0074] Based on the mapping table, rename the extracted original field names to the target field names;

[0075] Perform data cleaning: unit unification (convert units to standard units), coordinate system transformation (convert drawing coordinates to actual geographic coordinates if needed), and handling missing values ​​(fill missing fields with default values ​​or perform inference).

[0076] (1.2.4) Structured assembly:

[0077] The cleaned and aligned data is then assembled according to a predefined JSON Schema;

[0078] Ensure that geometric data uses a common structural representation (such as the geometry format).

[0079] (1.2.5) Output and Validation: Write the assembled data into a JSON file or store it directly in the database;

[0080] Perform data validation: format validation (ensure JSON conforms to the pre-designed schema), logic validation (check geometric validity (e.g., whether polygons are closed, whether required fields are complete), and visual spot checks (render the generated JSON in a simple front-end (e.g., Leaflet map, Three.js scenario), compare it with the original image, and ensure that the extraction is correct).

[0081] (1.2.6) Process automation: The above steps (i.e., the contents of steps 1.1 to 1.5) are written as scripts or pipelines to support batch processing.

[0082] (2) The task orchestration engine in the large language model receives the standardized data and starts the task parsing process. Based on the stratigraphic layer number in the standardized data, the report template selected by the user, and the complexity matrix, the task orchestration engine plans an execution sequence containing 20-80 atomic steps. That is, the task orchestration engine dynamically decomposes the exploration report to be generated into an execution sequence of 20-80 executable atomic steps.

[0083] In practice, the difference in the number of atomic steps is based on the following dimensions:

[0084] (2.1) Number of stratigraphic layers: If the site has a simple stratigraphic structure (e.g., only 3 layers), the profile generation only requires about 3-5 atomic steps of "reading data - calculating coordinates - receiving the connection". If the stratigraphic structure is complex (e.g., 15 layers, with lenses, faults or pinch-outs), then steps such as "lens closure processing", "fault line interpolation", and "stratigraphic pinch-out logic judgment" need to be added, which leads to a significant increase in the number of atomic steps.

[0085] (2.2) Report template type: The preliminary exploration report only requires basic logging and statistics (about 20-40 atomic steps), while the detailed exploration report needs to include advanced calculation steps such as liquefaction judgment, corrosivity evaluation, and pile foundation parameter calculation (about 60-80 atomic steps).

[0086] (2.3) Complexity matrix: The complexity matrix is ​​closely related to the number of attached plots. For each additional cross-sectional plot or bar chart, the task parser will automatically add a corresponding sub-sequence of "data extraction - coordinate transformation - plotting instruction generation".

[0087] In some embodiments, during the execution of each atomic step, a state manager manages the dependencies of each atomic step and tracks the execution status in real time. If a step fails, a rollback controller is used to roll back to the nearest stable checkpoint. In specific implementations, a directed acyclic graph (DAG) is used to manage atomic step dependencies.

[0088] The rollback controller primarily addresses the unreliability and state pollution issues inherent in Large Language Model (LLM) applications. Unreliability refers to the possibility that the LLM might output incorrect instructions, invoke incorrect tools, or generate unresolvable content. State pollution occurs when an erroneous output from one step is carried over to subsequent steps, leading to error accumulation and eventual task failure. Therefore, the rollback controller is used to restore the system state to a previously known correct point and attempt alternative paths when task execution deviates from expectations or encounters errors.

[0089] In practical implementation, the rollback controller mainly includes an execution monitor, a fault diagnostician, a rollback decision maker, a state restorer, and an alternative strategy executor. The rollback controller is tightly integrated with the task orchestration engine and the state manager. Specifically, the execution monitor is used to monitor the execution results of each step in real time; the fault diagnostician analyzes "what error occurred" and "why the error occurred"; the rollback decision maker, based on the diagnostic structure and predefined policies, decides whether to roll back, to which step, and how to retry; the state restorer collaborates with the state manager to execute the specific rollback operations; and the alternative strategy executor is used to adopt a different execution path after the rollback.

[0090] (3) Single-step generation loop: The large language model performs closed-loop processing for each atomic step.

[0091] In some embodiments, the large language model performs closed-loop processing on each atomic step, specifically including:

[0092] (3.1) The big oracle model constructs a context based on the execution structure of the preceding steps and uses the generation module in the big language model to perform parallel sampling to generate 3-8 candidate JSON instructions.

[0093] Ultimately, the user only needs a single, explicit action instruction. However, to ensure the correctness of this instruction, the large language model generates 3-8 candidate actions in parallel for the same prompt. In practice, the Temperature parameter generated by the large language model is set to 0.5-0.7 to introduce a slight degree of randomness and generate multiple possible JSON instructions.

[0094] (3.2) The generated instructions are immediately passed through the exception filter (red flag mechanism), which removes any candidates with format errors or values ​​that exceed the specification boundaries.

[0095] Each of the 3-8 candidate actions (i.e., 3-8 candidate JSON instructions) is checked independently. The red flag mechanism principle is as follows: once a candidate action triggers a rule red line (such as the JSON format cannot be parsed, the required field target_layer is missing, the generated coordinate value exceeds the site boundary, or it contains rejection keywords such as "I don't know"), the candidate action (i.e., the JSON instruction) is marked as "red flag" and directly eliminated. Only compliant candidates are retained to enter the next voting stage.

[0096] (3.3) For the remaining compliant candidate actions (i.e. JSON instructions), the robust voting device executes the strategy of "stopping when the first K votes are reached" by statistically analyzing semantic redundancy, and quickly selects the optimal action (i.e. the optimal JSON instruction) with the highest consensus.

[0097] In its implementation, the robust voting system is based on the principle of "semantic consistency." Under high probability, the same answers repeatedly generated by a large language model (i.e., a large model) are often the correct answers. A threshold K is set (e.g., K=3), and the similarity between the remaining qualified candidate answers is calculated (for JSON commands, key-value exact matching is used). During the generation or comparison process, once the number of times a certain answer content appears repeatedly reaches the threshold K (e.g., K=3, meaning it gets 3 votes first), the system immediately terminates subsequent sampling or comparison, directly determining that answer as the "optimal action." This ensures accuracy while avoiding the waste of computational power waiting for all 8 sampling cycles to complete.

[0098] (4) The optimal action is verified by a small model, and the verification is completed from at least two dimensions: compliance with engineering specifications and consistency with geological logic. If the verification is successful, the instruction is issued and executed; if it fails, a rollback and retry are triggered. This ensures the absolute reliability of each step.

[0099] Among these, Encoder-only architecture models (such as BERT, RoBERTa, or DistilBERT) with relatively small parameter counts (e.g., 110M-300M parameters) and extremely fast inference speeds are used for discriminative validation. These models excel at classification tasks and logical implication judgments.

[0100] In the specific implementation process, the optimal action is verified by a small model, which specifically includes:

[0101] Structure validation: Use Schema Validator to check the integrity of JSON structure, and use small models to determine whether data fields conform to the semantic definition of the current step (e.g., check whether the "stratum description" action includes features of three dimensions: color, density, and inclusions).

[0102] Compliance verification: Based on the Retrieval Enhanced Generation (RAG) approach, the numerical value in the optimal action (e.g., "cohesion = 5 kPa") is compared and verified with the provisions in the built-in standard database (e.g., the "Code for Investigation of Geotechnical Engineering" GB50021). The small model acts as a binary classifier, taking [standard provision, action parameter] as input and [compliant / non-compliant] as output. For example, if the standard stipulates that the void ratio of soft soil must be greater than 1.0, while the optimal action is 0.8, it is judged as non-compliant.

[0103] Logical consistency verification: Based on Natural Language Reasoning (NU), it checks whether the contextual logic is contradictory. For example, given the input pair: Premise: "The previous layer is the 3rd layer of silty clay", Hypothesis: "This layer is the 2nd layer of silt". According to the geological sedimentary pattern (older at the bottom, younger at the top), the small model will identify a logical contradiction in the stratigraphic sequence (the 2nd layer should not appear directly below the 3rd layer), thus rejecting the action.

[0104] (5) Verified instructions drive atomic executors to operate. If execution is successful, the global state pool is updated; if execution fails or verification fails, the corresponding rollback strategy is immediately initiated.

[0105] (6) Result Synthesis Output: After all atomic steps are completed, the output synthesis model collects all execution results, arranges the content and optimizes the format according to the exploration report annotation template, and generates the final exploration report.

[0106] The automatic subtask orchestration method for survey report generation in this invention uses a multi-step decomposition principle (that is, decomposing the complex survey report generation into multiple atomic steps). Each step has clear input, processing and output definitions, and the steps are connected through data dependencies.

[0107] Single-step generation principle: Large models generate only one atomic step's action description at a time, avoiding the error propagation problem in traditional chained generation. Each action uses a structured description to ensure parsing and execution. This ensures that the single-step output is highly consistent with engineering specifications.

[0108] The principle of multi-candidate selection: Multiple candidate actions are obtained through multiple samplings. After anomaly filtering, a voting mechanism based on multiple indicators is used to select the optimal solution, which significantly improves the reliability of single-step decision-making.

[0109] Real-time verification principle: A lightweight discrimination model is introduced, and RAG technology is used to search the survey specification database in real time to conduct engineering logic and compliance reviews on candidate actions.

[0110] Error isolation principle: By using atomic executors and rollback mechanisms, local execution exceptions can be captured in time and targeted retries can be triggered, ensuring the robustness of the global task chain.

[0111] In summary, based on the multi-step decomposition principle, single-step generation principle, multi-candidate selection principle, real-time verification principle, and error isolation principle of this invention, this invention has the following specific advantages:

[0112] 1. Significantly improved reliability: Error propagation is effectively suppressed, and the overall logical consistency of the survey report is greatly improved.

[0113] 2. Improved accuracy: Single-step output maintains a high degree of consistency with engineering specifications and experience-based structures.

[0114] 3. Significantly enhanced consistency: Graphs, tables, and queries are generated from the same source, and data with the same name is automatically aligned.

[0115] 4. Overall efficiency leap: The entire process time is significantly shortened compared to the traditional chain method, and manual operation is significantly reduced.

[0116] 5. Stability is fully guaranteed: the output fluctuation is small after multiple runs, and the system can automatically roll back and continue running in case of failure. Example 1 - Cloud Deployment

[0117] A container cluster is set up on a public cloud or industry private cloud. The data access module, task orchestration engine, large model generation module, discrimination and verification module, and atomic executor all run as microservices. Users drag and drop field records and CAD drawings into a browser → the backend automatically breaks them down into 60 atomic steps → each step is written by the large model with only one JSON action → the discriminator performs standard verification within 0.1 seconds → drives the execution of the existing "Lizheng + CAD + Excel" cloud container → write back on success and roll back on failure → download the entire report in 30 minutes. No software installation is required on the user side, and the original professional algorithms and mapping kernel are completely reused. Example 2 - Construction Site Integrated Machine

[0118] The system utilizes a 2U industrial chassis with an embedded domestically produced GPU. It comes pre-installed with large language models, discrimination and verification models, and executable programs such as LiZheng, CAD, and Office. The process involves plugging in the box, importing data from a USB drive, local disassembly, gradual generation of a 32B model within the domestic GPU, real-time discrimination and scoring of small models, calling the pre-installed LiZheng / CAD / statistics executables, and automatically printing a PDF and burning it to a read-only CD. No data is left on the machine throughout the entire process, and a signed survey report is output. Example 3

[0119] 1. System Deployment and Configuration: Deploy the large language model and the discriminant verification model (small model) on the server; build a historical task knowledge base to store atomic action templates and specification constraints; configure the API interfaces of various professional software.

[0120] 2. Task initialization: Upload raw exploration data (drilling data, test results, site information, etc.); select report template and output specification requirements; set quality acceptance standards and mandatory clause checklist.

[0121] 3. Intelligent orchestration and execution: The large model (i.e., large language model) parsing module breaks down the overall task into a sequence of atomic steps; the multiple sampling module generates multiple candidate actions for each step; the discrimination and verification model checks the standard compliance of the candidate actions; the voting selection module determines the optimal action based on multi-index evaluation; and the atomic executor calls the corresponding professional software to execute the verified action.

[0122] 4. Quality monitoring and exception handling: The execution engine monitors the execution status of each step in real time; when a failed step is encountered, it automatically rolls back to the previous checkpoint; the generation-verification-execution loop of that step is retried; error logs are recorded and the knowledge base is updated to avoid repeated errors.

[0123] 5. Results Output and Delivery: The output module summarizes the execution results of all successful steps; automatically generates a Word report document that conforms to the specifications; packages and outputs CAD drawings, statistical tables and other supporting materials; and generates a quality control report indicating the pass status of key inspection items.

[0124] In summary, this invention, through an innovative architecture of multi-step decomposition, discrimination verification, and robust voting, achieves intelligent arrangement and stable execution of the entire process from "data input - stratigraphic division - parameter statistics - drawing generation - text writing - report synthesis." It effectively solves industry pain points in existing technologies such as "unreliable large model generation, cross-step error propagation, and difficulty in implementing standard provisions." While maintaining the existing professional software system, it significantly improves the automation level and engineering reliability of exploration report generation, demonstrating significant technological breakthrough value and promising prospects for widespread application.

[0125] It should be understood that the above description of the preferred embodiments is quite detailed, but it should not be considered as a limitation on the scope of protection of this invention. Those skilled in the art, under the guidance of this invention, can make substitutions or modifications without departing from the scope of protection of the claims of this invention, and all such substitutions or modifications fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.

Claims

1. A method for automatically arranging sub-tasks for a survey report generation task, characterized in that, include: (1) Access multi-source heterogeneous data and perform standardization processing. The standardization processing converts multi-source heterogeneous data into a unified structured JSON format, thereby obtaining standardized data; (2) The task orchestration engine receives standardized data and starts the task parsing process. Based on the number of stratigraphic layers in the standardized data, the report template selected by the user, and the complexity matrix, the task orchestration engine plans an execution sequence containing 20-80 atomic steps. (3) Single-step generation loop: The large language model performs closed-loop processing for each atomic step; (4) The optimal action is judged and verified by a small model, and the verification is completed from at least two dimensions: engineering specification compliance and geological logic consistency. If the verification is successful, the instruction is issued and executed; if it fails, a rollback and retry are triggered. (5) Verified instructions drive atomic executor operations: if the execution is successful, the global state pool is updated; if the execution fails or the verification fails, the corresponding rollback strategy is immediately started. (6) Result Synthesis Output: After all atomic steps are completed, the output synthesis model collects all execution results, arranges the content and optimizes the format according to the exploration report annotation template, and generates the final exploration report.

2. The method of claim 1, wherein the method further comprises: The difference in the number of atomic steps is based on the following dimensions: (2.1) Number of stratigraphic layers: If the site has a simple stratigraphy, it takes about 3-5 atomic steps; if the stratigraphy is complex, it requires atomic steps. (2.2) Report Template Types: Report template types are divided into preliminary survey reports and detailed survey reports; (2.3) Complexity matrix: The complexity matrix is ​​related to the number of attached diagrams. For each additional cross-sectional diagram or bar chart, the task parser will automatically add a corresponding subsequence.

3. The method of claim 1, wherein the method further comprises: During the execution of each atomic step, the state manager manages the dependencies of each atomic step and tracks the execution status in real time; if an atomic step fails, the rollback controller is used to roll back to the nearest stable checkpoint.

4. The method of claim 1, wherein the method further comprises: The large language model performs closed-loop processing on each atomic step, specifically including: (3.1) The big oracle model constructs a context based on the execution structure of the preceding steps and uses the generation module in the big language model to perform parallel sampling to generate 3-8 candidate JSON instructions; (3.2) The generated instructions are immediately passed through the exception filter, which removes any candidates with format errors or values ​​that exceed the specification boundaries; (3.3) For the remaining compliant candidate actions, the robust voting device executes the strategy of "stopping when the first K votes are reached" by statistical semantic repetition, and quickly selects the optimal action with the highest consensus.

5. The method of claim 1, wherein the method further comprises: The optimal action is verified by a small model, including structural verification, specification compliance verification, and logical consistency verification.

6. The method of claim 5, wherein the method further comprises: The structure verification includes: using a Schema Validator to check the integrity of the JSON structure, and using a small model to determine whether the data fields conform to the semantic definition of the current step.

7. The method of claim 5, wherein the method further comprises: The compliance verification includes: based on the retrieval enhancement generation approach, comparing and verifying the value in the optimal action with the provisions of the built-in specification database.

8. The method of claim 5, wherein the method further comprises: The logical consistency verification includes checking whether the contextual logic is contradictory based on natural language reasoning.

Images