56 results about "Task analysis" patented technology

The application discloses a cross-terminal task processing method and device, electronic equipment, medium and program product. The method comprises the following steps: obtaining task indication information corresponding to a to-be-processed task, wherein the task indication information is conversation information input by a current account on a first terminal and directed to a preset intelligent agent; performing task analysis on the task indication information based on the preset intelligent agent to obtain a task instruction sequence, wherein the task instruction sequence is an operation instruction in a process of executing the to-be-processed task based on a target application, and the target application is an application program corresponding to a second terminal; deploying the target application in a local emulator instance to obtain an emulator instance carrying the target application; and executing the task instruction sequence based on the emulator instance carrying the target application. The technical scheme provided by the application can improve the task processing efficiency and convenience, effectively cope with different cross-terminal task processing requirements, and improve the efficiency and interaction convenience of cross-terminal collaborative processing.

This invention relates to the field of inventory management technology, and proposes a method and system for real-time monitoring of material inventory based on Internet of Things (IoT) sensing. The method includes: distributing personalized data collection strategies to IoT sensing nodes via a communication network of a central monitoring platform; performing differentiated processing on initial inventory sensing data based on these strategies to generate inventory status records; dynamically adjusting the reporting frequency of status priority tag sequences and sending the adjustment process and records to the central monitoring platform in real time; correlating and analyzing the inventory status records with real-time sensing signals to construct a multi-dimensional dynamic inventory ledger; comparing the real-time inventory quantity in the ledger with a safety stock threshold, and performing task analysis in conjunction with an anomaly triggering mechanism to obtain an adaptive task list. This invention can improve the real-time performance, accuracy, and adaptive efficiency of material inventory monitoring.

The application discloses a kind of unmanned aerial vehicle real-time semantic perception systems and methods based on flight state and task adaptation.The system includes flight state perception module, task analysis module, adaptive scheduler and configurable perception engine.The core is, according to unmanned aerial vehicle real-time flight state (such as attitude, maneuver mode) and preset task stage (such as search, crossing, landing), dynamically scheduling sensor configuration, selecting algorithm model and adjusting semantic output granularity.The application solves the problem that the resource consumption of existing unmanned aerial vehicle perception system is fixed and the environmental adaptability is poor, realizes the optimal matching of perception ability and task demand, significantly improves the efficiency, safety and overall system robustness of unmanned aerial vehicle autonomous flight in complex dynamic environment.

The present application relates to the technical field of automation plate processing and carrying control, and particularly relates to a plate automatic feeding and discharging control system and a control method thereof, which comprises: a task analysis module, which receives an external production task and calculates the external production task into processing procedure parameters containing plate geometric size, thickness and material density characteristics; and a gravity calculation module, which calculates plate mass distribution based on the processing procedure parameters and outputs gravity matrix data containing target gravity coordinates, suction cup matrix gating strategy and plate specification characteristics. In the present application, through automatic detection and calculation of plate geometric characteristics and density information, on-demand dynamic gating of the suction cup array and adaptive calculation of the vacuum negative pressure threshold are realized, frequent parameter adjustment caused by plate specification changes in the traditional manual teaching mode is avoided, and the automatic adaptation capability of the feeding process to multi-specification plates is improved.

The application relates to a robot safe operation method and system based on a double-system architecture, and belongs to the field of robot operation. The method comprises the following steps: task analysis is performed on a natural voice instruction, and a task in the natural voice instruction is decomposed into multiple subtasks; for each subtask, target object point cloud and observation scene point cloud are determined according to robot vision information; an optimal grabbing position and an optimal grabbing posture are determined based on a grabbing posture generation network according to the target object point cloud and the observation scene point cloud; a path map, a posture map and a gripper state map are constructed based on the optimal grabbing position and the optimal grabbing posture; path planning is performed according to the path map, and posture planning is performed according to the posture map, so that a final trajectory is obtained; and the robot is controlled according to the final trajectory and the gripper state map to complete the subtask. The application combines the semantic understanding ability with the execution robustness of a traditional planning control method, and realizes task-driven operation without samples and training.

The application discloses a spectrum detection method based on a large language model, applied to the technical field of spectrum detection, and combines field instruction fine-tuning and retrieval enhancement generation technology, and introduces a parameter efficient fine-tuning method such as LoRA, effectively solves the problems of insufficient understanding of scene problems, model illusion and high training cost of general models in professional fields. With the physical constraint spectrum data augmentation algorithm and the specially designed double task analysis network, the feature extraction and quantitative analysis ability of small sample and low concentration spectrum data is improved. The finally constructed model of the application has the advantages of professional question and answer accuracy, reliable result and light deployment, and only needs a single consumer-grade GPU to complete training and deployment, providing an efficient, reliable and low-cost intelligent auxiliary analysis solution for spectrum detection.

The application discloses a business auditing method and system architecture based on a large language model and a dynamic DAG arrangement, and relates to the technical field of artificial intelligence and task automation processing. The method is applied to a large language model and includes the following steps: receiving a to-be-audited file and auditing requirement data; performing semantic intention understanding and task disassembly processing on the auditing requirement data to determine atomic tasks; analyzing the dependency relationship between the atomic tasks and generating an initial DAG task graph based on the dependency relationship, wherein the initial DAG task graph includes multiple mutually dependent nodes; determining a current executable node from the initial DAG task graph; executing the current executable node, calling an auditing tool and / or model from a multi-modal tool library to perform business auditing on the to-be-audited file to obtain an auditing result, wherein in the case where the current executable node includes multiple parallel nodes, the multiple parallel nodes are executed in parallel. Through the application, the business auditing efficiency, flexibility and adaptability are effectively improved.

The specification discloses a data transfer service test method and device, a storage medium and an electronic equipment. The method is applied to the field of computer technology and comprises the following steps: firstly, determining a data transfer service test instruction for a data transfer service; then, based on the data transfer service test instruction, using a test task analysis intelligent agent to perform knowledge retrieval processing in a data transfer service knowledge graph to obtain service test associated knowledge, and generating a service test action call chain based on the service test associated knowledge and the data transfer service test instruction; then, using a test action execution intelligent agent to sequentially determine a current service test action from the service test action call chain, assemble service test environment data of the current service test action, execute the current service test action based on the service test environment data to obtain a service test execution result; and finally, generating a data transfer service test report based on the service test execution result.

The application provides a full-process dynamic task planning AI agent scheduling method and system for SEO, relates to the technical field of agent scheduling, and preconfigures rule configuration information after building an agent core architecture, connects the preconfigured rule configuration information to a plurality of LLM models and a full-network information collection interface, and obtains preparation architecture building information; the preparation architecture building information is used for creative task analysis through a retrieval hub, creative task analysis data is obtained, the creative task analysis data is analyzed for task allocation through the retrieval hub, task allocation analysis data is obtained; the keywords, outline and content are sequentially generated and verified and analyzed through the multi-agent architecture building information, and agent scheduling data is obtained, the application effectively solves the contradiction between the plurality of LLM models and task adaptation and the problem of scheduling hub computing power overload under high concurrency, breaks the vicious circle of the two, does not require manual intervention, and greatly improves the full-process automation level of SEO content creation.

In order to solve the resource problem in the heterogeneous cloud computing environment, the application provides a batch processing task adaptive scheduling method based on pulse reinforcement learning. The technical path covers: multi-dimensional feature decoupling extraction is carried out on the input batch processing task, and clustering analysis is carried out; task classification is realized based on task characteristic attributes, and sorting is carried out according to priority; based on task analysis modeling, a pulse reinforcement learning resource scheduling model is established, and a double sparse attention mechanism is introduced to realize global state updating; the task scheduling environment of pulse reinforcement learning is constructed; the decision network and evaluation network of the pulse reinforcement learning model are constructed, and finally, in the simulation simulation scene, the trained model is used to realize the adaptive scheduling of batch processing tasks. The application fully considers the characteristic attributes of different tasks, overcomes the performance bottleneck of traditional scheduling algorithms in complex heterogeneous environment through accurate supply and demand matching strategy, and reduces the inference energy consumption during model training.

The present application relates to the technical field of stereoscopic storage automation, in particular to a stereoscopic storage multi-axis linkage intelligent goods shelf position dynamic allocation and scheduling system, which comprises a task analysis module, an environment perception module, a goods position recommendation module, a scheduling optimization module and a plan execution module. The task analysis module receives and analyzes various storage operation instructions in real time. The environment perception module dynamically perceives the real-time state of the whole warehouse area. The goods position recommendation module generates multiple candidate goods position sets for the goods to be stored according to the goods attributes and storage requirements, in combination with the real-time state of the whole warehouse area, by using an improved collaborative filtering algorithm based on the correlation of goods storage characteristics. The scheduling optimization module evaluates each candidate set according to the preset optimization target, selects the final goods position, and generates a detailed storage execution plan containing the collaborative operation path of multiple tunnel stacker cranes. The plan execution module schedules the equipment execution plan and monitors and updates the state in real time.

This invention relates to the field of container terminal management technology, specifically a dynamic task planning system for yard cranes under multi-factor constraints. The system includes a yard crane data acquisition module, a yard crane multi-task analysis module, a yard crane information processing module, and a yard crane collaborative scheduling module. Based on the errors and influencing factors of the data acquisition module, the sampling time is adjusted to obtain a yard crane monitoring database. Based on the yard crane task time quantification analysis system of the yard crane multi-task analysis module, time analysis results and multi-task comprehensive analysis results are obtained for different task stages during yard crane operations. The yard crane information processing module classifies the database to obtain data subsets for different task stages during operations. The yard crane collaborative scheduling module combines the time analysis results and data subsets of different task stages with the multi-task comprehensive analysis results to adjust and plan yard crane tasks. This invention's data-driven dynamic scheduling method enables the system to adapt to different operational scenarios and task requirements, enhancing system adaptability and stability.

The application provides a structured outline driven scientific report automatic generation method and system, and relates to the technical field of natural language processing, the method comprises the following steps: performing semantic analysis and matching on a user input research topic to generate a first-level structure node, extracting and dividing research problems from multiple standard scientific literatures to construct a second-level structure node, performing opinion sentence extraction and refinement to generate a third-level structure node, generating a structured outline according to the first-level, second-level and third-level structure nodes and corresponding explicit binding relationships, performing task analysis, generating a task set according to the hierarchical order of the structured outline, generating a target scientific report and outputting the target scientific report by performing text generation, assembling and integrating. The technical problems of unstable structure, easy divergence of theme, difficulty in corresponding paragraphs and evidence, and easy repetition of chapter content in the existing scientific report automatic generation are solved. The technical effect of stable structure, focused theme and traceable content of the scientific report automatic generation is achieved.

This invention discloses a multi-level human-computer interaction intent recognition method for complex task scenarios, belonging to the field of human-computer interaction intent recognition technology. It solves the technical problems of low accuracy and low interpretability of multi-level human-computer interaction intent recognition in complex task scenarios. The key technical solution proposes a multi-level intent recognition framework suitable for complex task scenarios, combining task analysis methods and machine learning methods. This framework possesses both sensitivity to human behavior and contextual information, as well as interpretability. Its input data consists of easily captured human behavior data and situational data that can be collected by machine sensors, offering the advantages of not interfering with the working process of the human-computer system and its operability. By simultaneously integrating human behavior, environmental, and task context features, it identifies multi-level human intents layer by layer from the bottom-level data, achieving a more comprehensive understanding of the true human intent. This effectively overcomes the limitations of existing methods that rely on a single level and only on human behavior characteristics for intent recognition, improving the accuracy of multi-level intent recognition while simultaneously achieving interpretable modeling of multi-level intents.

The application discloses a kind of based on cloud edge collaborative processing's internet of things information management system, belong to internet of things information management technical field, by real-time collection internet of things data to sensing terminal, and transmit internet of things data to edge end;Edge end is preprocessed to the internet of things data received, obtains standardized internet of things data;Real-time determination edge analysis task, according to edge analysis task call corresponding built-in task analysis mode is analyzed to standardized internet of things data, obtains edge task analysis result;Standardized internet of things data and edge task analysis result are screened, obtain high-value data, abnormal data and data abstract, and high-value data, abnormal data and data abstract are sent to cloud end;Cloud end is according to the high-value data, abnormal data and data abstract received task analysis processing;And sensing terminal is collected parameter management.

The present application relates to the field of work ticket supervision, and particularly relates to a power work ticket intelligent supervision system fusing multi-modal data, comprising: a data analysis unit, used to determine whether to perform high-precision optimization for a benchmark verification mode based on a split condition; a parameter optimization unit, used to determine whether to adjust path enumeration depth based on a regional topology complexity corresponding to a regional condition and a dynamic load reference degree; a processing optimization unit, used to determine whether to perform path order optimization based on a redundancy condition; and a task analysis unit, used to determine whether to adjust a cache update duration or perform task flow conversion based on a range conflict value corresponding to a processing module, and determine a flowable module based on a range discrete value in the task flow conversion. The present application improves the supervision efficiency of the power work ticket.

Embodiments of the present application disclose a RHEED image multi-task analysis method and device, electronic equipment and storage medium, relating to the technical field of computer vision and semiconductor preparation process monitoring, wherein the method comprises: constructing a training data set, using ConvNeXt V2 designed in cooperation with a full convolution mask autoencoder and a global response normalization layer as a backbone network, combining a multi-task learning architecture, setting classification, detection and time series prediction branches to construct an analysis model. In the single-task training stage, parameters are adjusted, self-supervised pre-training parameters are introduced, and other optimizations are performed for each task; in the multi-task cooperative training stage, the parameters are frozen, gradually unfrozen, and the overall objective function is set. After preprocessing the to-be-tested data, input the model, extract multi-scale features and perform multi-task branch reasoning to obtain classification results, target detection results and RHEED intensity oscillation curve prediction values. The present application effectively solves the problem that the prior art is highly subjective and difficult to adapt to different equipment and scenes.

The application relates to the technical field of cloud platform resource management, in particular to a cloud platform resource intelligent management system and method based on distributed edge computing. The system comprises a resource sensing unit, a task analysis unit and an intelligent scheduling unit. The intelligent scheduling unit is used for realizing optimal matching of resources and tasks, generating resource allocation instructions based on resource dynamic data and task characteristic parameters by using a multi-objective optimization algorithm with dynamic weight adjustment combined with load balancing logic. The system further comprises an execution monitoring unit. The hardware running state and software resource configuration of edge nodes and the cloud platform are collected and pretreated in multiple dimensions by the resource sensing unit, the task characteristics are quantitatively analyzed by the task analysis unit, and the multi-objective optimization algorithm with dynamic weight adjustment based on task priority and node load difference is used by the intelligent scheduling unit, so that the matching and adaptation degree of resources and task requirements is improved, the resource allocation is more suitable for actual task scenarios, and the situation of resource mismatch or redundant allocation is reduced.

The application discloses a one-stop intelligent software development system based on multiple agents, which comprises a one-stop DevOps research and development cooperation platform and a multiple-agent scheduling and cooperation engine. The multiple-agent scheduling and cooperation engine is connected with each sub-module in the one-stop DevOps research and development cooperation platform, including a demand management module, a code library management module, a document management module, a pipeline module, a product library module, a code scanning module, a test management module and a measurement management module. The multiple-agent scheduling and cooperation engine comprises a task analysis and distributor, a multi-role virtual employee intelligent agent pool, a large model calling adapter, a delivery material archiving and integration controller and an artificial confirmation node manager, realizes end-to-end automation from demand description to software delivery, and only needs to fill in a natural language demand description or a prompt word in a task card, so that the system can automatically drive a whole-process task such as prototype generation, code writing, test case generation and construction deployment, and the software delivery cycle is greatly shortened.

This application discloses a data processing method and related apparatus. The method involves acquiring a task to be analyzed, which identifies the target operation of an account in a data system, the corresponding operation log, and the security operation regulations involved in the target operation. A first large-scale model is instructed to decompose the task into N progressively smaller subtasks from the perspective of operational security review. The granularity of the analysis dimension decomposition of the i-th subtask is greater than that of the (i-1)-th subtask, where N>1. The first large-scale model is then instructed to perform chain-like analysis on the N subtasks in a progressive order to obtain task analysis results. These results indicate whether the target operation triggers security operation regulations in the data system, as described in the operation log. This application improves the efficiency and accuracy of security analysis by performing chain-like analysis based on a coarse-to-fine decomposition granularity when using a large-scale model for security analysis.

The application relates to a method and device for analyzing key tasks in earthquake rescue and a computer device. The method comprises the following steps: acquiring a plurality of rescue tasks of an earthquake rescue operation, the duration of each rescue task, and the type and quantity of rescue resources consumed by each rescue task; constructing an initial task coordination diagram according to the execution logic of the plurality of rescue tasks and the duration of each rescue task; performing resource conflict detection of reusable resources and non-reusable resources on the initial task coordination diagram according to the type and quantity of rescue resources consumed by each rescue task, and adjusting the initial task coordination diagram to obtain a feasible task coordination diagram according to the resource conflict detection result; and obtaining key tasks in earthquake rescue according to pre-set time key indicators and degree key indicators. The method can analyze key rescue tasks from tasks with multiple levels, a large quantity and complex dependency relationships, and can provide key support.

The application is suitable for the technical field of medical devices, and provides an interventional instrument distal end force prediction method, device and program product. The method comprises: acquiring image data of an interventional instrument in a biological cavity; inputting the image data into a prediction model; wherein the prediction model comprises an encoder and a decoder; extracting a first feature containing interventional instrument deformation information from the image data based on the encoder; processing the first feature based on the decoder to obtain an attention weighted feature focusing on a distal end deformation region of the interventional instrument; performing multi-task analysis on the attention weighted feature based on the decoder to output a distal end contact force prediction value and a distal end contact state level. The application realizes non-contact distal end force measurement by constructing an end-to-end prediction model from image to mechanical state, and improves the accuracy and robustness of prediction through the attention mechanism and multi-task analysis, thereby providing comprehensive mechanical state information for a surgical robot and improving the safety of surgery.

The application discloses a task-aware-based large model fine-tuning method and a legal information analysis method. The fine-tuning method comprises the following steps: constructing a target field knowledge graph and injecting an initial large model to obtain a first large model, establishing a mapping relationship between a task type and a knowledge subgraph to obtain a second large model, and configuring a task self-adaptive optimization mechanism to obtain a third large model; selecting and expanding initial data samples to obtain target training samples; finally, the third large model is fine-tuned by using a progressive unfreezing strategy and mixed precision training. The method deeply couples the two, uses a task-knowledge mapping table to drive the unfreezing sequence and dynamically update the precision bit width, converges quickly with the least video memory and parameters in the early stage, gradually releases the capacity and precision bit width in the later stage, realizes the triple balance of gradient stability, video memory saving and performance optimization, and the fine-tuned large model realizes the deep fusion of professional field knowledge and parameters, can dynamically adjust the internal representation and reasoning mode according to different task characteristics, and meets the diversified task analysis demand.

The application discloses a kind of intelligent navigation method, navigation system and medical transfer car of medical transfer car, it is related to intelligent navigation technical field.The method includes: task initialization and parameter configuration, real-time multi-source fusion positioning, dynamic calibration and scene adaptation and safety control and navigation optimization.The application parses medical task and material information configuration initial parameter by task analysis module, adopts double-dimension weight algorithm and Kalman filter to fuse multi-source positioning data in combination with data fusion calculation module, is equipped with inertial measurement unit double benchmark calibration, environmental adaptation switching module scene parameter adaptation, vibration sensing module vibration control and action prediction module trajectory prediction, realize in-hospital whole scene accurate navigation, guarantee high-precision following during first aid and routine task equipment safety, simultaneously, through storage optimization module iteration optimization navigation parameter, overall improve transfer efficiency, reduce collision and material damage risk, meet the core demand of medical transfer safety, efficient.

The application discloses a kind of low-energy consumption chemical experiment AI intelligent scheduling system and method of fusion AGV feedback, system includes experiment task analysis and modeling module, multi-source heterogeneous sensing and data fusion module, intelligent scheduling and dispatch kernel module, interactive scheduling confirmation and visualization module and instruction issue and dynamic feedback execution module.The application realizes the deep coupling of chemical process logic and AGV dynamic feedback by directly incorporating AGV physical operating characteristics into scheduling constraints;Through the construction of improved NSGA-III multi-objective optimization model with total completion time, AGV total transportation energy consumption and chemical timeliness satisfaction as the target, and the introduction of dynamic penalty mechanism based on health degree, the efficiency, energy efficiency and timeliness are balanced organically;Through the double-layer architecture of global optimization and local rescheduling and closed-loop dynamic feedback execution mechanism, the system can automatically trigger local rapid replanning when facing AGV blockage, low battery and other abnormalities.

This invention relates to the field of project management knowledge graph technology, and more particularly to an intelligent project collaboration platform for engineering management. It constructs a supervisory role knowledge graph through a role graph construction module, decomposes real-time power supervision requirements and constructs a real-time task knowledge graph through a task graph construction module, merges the two graphs through a knowledge graph fusion module and constructs a role-task fusion graph, then uses a collaborative task analysis module to obtain the aggregated result using a graph neural network model, and finally obtains a role-task allocation scheme through a role-task allocation module. This invention constructs a knowledge network linking "people" and "tasks" through a knowledge graph, and finally uses a graph neural network model for information aggregation to obtain an allocation scheme that balances optimal personnel capabilities, reasonable task dependencies, and the highest collaborative efficiency. This effectively solves the problem of broken collaborative chains in multi-disciplinary cross-task allocation faced by existing power supervision tasks.

The application discloses a kind of multi-agent collaborative distributed task optimization and learning system, it is related to collaborative task optimization technical field.System includes: task analysis module obtains computation graph;Soft boundary modeling module constructs the soft boundary state vector including split tendency value, state confidence and direction inertia value for directed edge, based on the ratio of operator calculation amount and output data volume is corrected initially and segmented task fragment;State perception module collects local real-time running index to calculate adaptation degree score, determines bottleneck position by trend analysis and attribution inference;Boundary adjustment decision module calculates migration potential energy to select adjustment target, executes split tendency value directional increase and decrease and updates state confidence and direction inertia value;Conflict resolution module arbitrates conflict adjustment based on weighted priority;Convergence control module freezes split scheme after meeting convergence condition and locally reactivates when environment changes.The application realizes the self-adaptive adjustment of distributed task split granularity at runtime.