41 results about "Decision points" patented technology

The invention provides a data processing method based on request interception, which comprises the following steps of: intercepting a calling request for marking a service method through a request interception mechanism, preposing a decision point of service logic, and selecting whether the service method is executed or not according to a request context. According to the method and the device, the business rules are stored in the configuration center, external rule configuration is performed, the business rule expressions are pre-stored in the configuration center, dynamic acquisition is performed based on the rule identifiers, decoupling of the business rules and the codes is realized, the business rules can be configured independent of the codes, and flexible business processing can be realized only by modifying the expressions of the corresponding rules in the configuration center. The problems of code expansion and difficult maintenance caused by business rule change in a multi-channel environment are effectively solved, so that the flexibility and maintainability of a business data processing system are improved. The method can be applied to a business processing system in the financial field or the medical field, so that the flexibility of the business processing system in the financial field or the medical field is improved.

The invention relates to the technical field of artificial intelligence, and discloses a military simulation agent design method based on a knowledge graph, and the method comprises the steps: taking a synchronous matrix, an execution matrix and decision support information formed in the military planning and war game deduction process as the basis, and uniformly representing units, stages, combat functions, decision points and condition elements in the knowledge graph, and a computable high-dimensional plan state expression is formed. On the basis, a plan updating model reflecting stage propulsion, function collaboration and decision point triggering relations is constructed and used for describing dynamic evolution of plan states in the simulation process. Through systematic description of a transient amplification phenomenon in a plan evolution process, clear association is established between the transient amplification phenomenon and decision points and condition elements in a knowledge graph, so that a simulation agent has stable and reproducible behavior performance in decision generation and action execution processes.

The application discloses a kind of based on blockchain's pedigree data dynamic permission access control system and method, rely on attribute-based access control paradigm, and combine pedigree data access constraint to carry out dynamic access control.First, user sends access request to policy decision point;Decision point according to the policy loaded from policy management point, request relevant information to blockchain, and call user historical behavior verification module based on pedigree data, the legality of current access request is verified using dependency relationship and pedigree data graph.System administrator records access request and verification result to blockchain, to support the fast verification of same query, reduce query overhead.In addition, the system passes access information such as query user, time, result, operation content to management node, for subsequent user access tracking and management.The method uses the anonymity and non-tamperability of blockchain, provides strong evidence for user supervision, realizes efficient management and reasoning to source information.

Systems and methods for implementation of a multi-agent reinforcement learning based decision system for a materials handling system, including initializing a simulation environment comprising decision points for dispatching materials and attributes of the materials handling system, the simulation environment configured to request a decision for materials dispatch at the decision points to the multi-agent reinforcement learning based decision system; initializing the reinforcement learning agents representative of the decision points for the multi-agent reinforcement learning based decision system; initializing domain expert heuristics for the decision points for the multi-agent reinforcement learning based decision system; and iteratively training the reinforcement learning based decision system with the initialized domain expert heuristics on the simulation environment.

The application provides a man-machine cooperative cognitive decision space construction and on-the-spot decision method, that is, in the action planning process, on the basis of the main line action plan designed based on the static task purpose, a person estimates possible variables in action execution in advance, refines situation cognitive characteristics of key decision points, and proposes countermeasures, the system explores diversified possible action processes, analyzes and excavates key variables which have greater influence on action results, explores the best action options under the guidance of countermeasures and trains a dynamic planning intelligent agent, finally, the cognitive characteristics and countermeasure options, the planning intelligent agent are added to the main line action plan, and a cognitive-decision space is constructed; in the action execution process, the cognitive-decision space is used to guide the real-time monitoring of the change of situation characteristics and the dynamic triggering of key decision points, the online optimization of the best countermeasure and the dynamic generation of subsequent action planning.

The invention provides a dynamic self-adaptive DNN collaborative reasoning method supporting a backspacing mechanism, relates to the technical field of image processing, and designs an elastic two-way task flow from equipment to edge to equipment, so that the elastic two-way task flow becomes an intelligent scheduling decision point. And when the edge server is overloaded, the task can be actively returned to the source mobile equipment to be executed, so that the bottleneck of queuing delay caused by edge resource competition is fundamentally avoided. A rollback decision is introduced, so that an unloading strategy is upgraded from'static 'or'semi-static' to real'dynamic self-adaption ', the change of edge resources can be directly responded, and a decision target is expanded from optimizing a single task to optimizing a global system load. Three key dimensions, namely a backspacing decision, resource allocation and a model segmentation point, are arranged in a unified framework for joint optimization. Through an integrated reinforcement learning model, the three decisions are output at the same time, so that the three decisions can cooperate with each other and have a synergistic effect.

This invention provides an industrial control method and system based on anthropomorphic thinking chain and hybrid expert model, including: Step S1: Based on the operator's thinking decision points and decision chain when controlling the equipment, construct a decision model based on anthropomorphic thinking chain and multi-layer hybrid expert model for each process segment in the complete production chain; Step S2: Train the decision model of each process segment using historical data to obtain the trained decision model; Step S3: Use the trained decision model to obtain the control setpoint based on the real-time sensing data of the corresponding process segment.

The invention discloses an intelligent generation method and device for a fine-grained data access control strategy and a medium, and belongs to the technical field of data security and artificial intelligence. S2, data processing; s3, generating strategy parameters, acquiring an initial data set required by clustering by utilizing the stored user information and the access log, and acquiring required strategy parameters by combining a DAC (Digital-to-Analog Converter) algorithm; s4, storing strategy parameters; s5, constructing a strategy mechanism; and S6, strategy use: calling the strategy through the strategy decision point, obtaining related parameters in the knowledge graph, returning an access request result to the strategy implementation point, performing access control operation by the strategy implementation point according to the result, providing data if passing, recording an alarm log while providing data if giving an alarm, and refusing access if blocking. And generating an alarm log. According to the invention, intelligent generation of the data security access control strategy is realized.

The invention discloses a network security operation behavior knowledge graph construction method and device, and belongs to the technical field of network security education. The method comprises the following steps: defining node types including operation steps, risk behaviors, safety principles, capability dimensions and decision points; defining relationship types of inclusion, time sequence, mapping, cause and examples; independently obtaining a standard operation process from the expert knowledge base, the attack and defense drill data or the user behavior log, and extracting steps, attributes and corresponding safety principles and capability dimensions of the standard operation process; mining a risk behavior pattern from the historical user behavior data; integrating the nodes and the relationships to construct a knowledge graph and storing the knowledge graph in a graph database; and establishing a mapping interface from the behavior sequence to the map for real-time behavior matching and capability diagnosis. The invention further provides a method and a device for user evaluation, group analysis and collaborative application with the event case knowledge graph (including but not limited to prior applied patents) based on the graph. According to the method, structured modeling is carried out on fragmented operation behaviors, accurate mapping of behaviors and capabilities is established, interpretable capability diagnosis and personalized teaching support are realized, and the method can be widely applied to network security awareness training and behavior analysis systems.

The present disclosure relates to a driving planning method, device and equipment of a vehicle and a storage medium. The present disclosure obtains driving decision information of the vehicle by analyzing the action trajectory of at least one obstacle according to the obtained dynamic information of the obstacle, wherein the driving decision information at least includes at least one driving decision point and a vehicle driving strategy corresponding to each driving decision point; performs tree search on the at least one driving decision point according to the driving decision information, plans the driving acceleration of the vehicle for the at least one driving decision point, and obtains the vehicle driving strategy according to the driving acceleration, so that the obtained vehicle driving speed is more accurate. Compared with the prior art, the present embodiment avoids returning to the initial position for decision planning when there is no vehicle driving strategy for the current obstacle, saves the time for planning the vehicle speed, and improves the planning efficiency.

The application discloses a multi-source data fusion and dynamic scheduling distributed intelligent deduction system and method, and relates to the technical field of deduction simulation. A deduction preprocessing module divides a military scenario into task stages and sets decision points, generates branch tasks according to tactical rules and classifies models; a deduction management module advances deduction from an initial state based on a simulation engine, generates branch tasks according to real-time states at the decision points, combines historical performance parameters and model types to distribute initial weight coefficients, realizes parallel deduction and dynamic model switching; a deduction optimization module continuously updates weight coefficients through branch performance parameters, filters optimal deduction paths and closes loop iteration state snapshots; a resource regulation module collects data in real time to support performance evaluation, dynamically optimizes resource allocation according to task load, significantly improves deduction efficiency and accuracy, and realizes intelligent resource scheduling and system adaptive optimization.

This invention discloses a method for calculating the takeoff trajectory of a helicopter in Category A flight. The key technical points include the following calculation steps: S1: Statistically analyze the parameters of a twin-engine helicopter and the parameters of the engines it uses; S2: Based on the statistically analyzed parameters, calculate the power required for forward flight of the twin-engine helicopter and plot N. xu -V0 curve; S3: Calculate the safe takeoff speed of the twin-engine helicopter based on the power required for forward flight and the statistical power of a single helicopter engine; S4: Calculate the lift and forward thrust provided by the helicopter rotor based on the statistical parameters; S5: Calculate the critical takeoff decision point of the twin-engine helicopter and plot the critical takeoff decision point curve; S6: Based on the calculated critical takeoff decision point of the twin-engine helicopter, calculate the takeoff trajectory of the twin-engine helicopter in Category A flight according to the location of the failure point, and plot the Category A takeoff trajectory curve of the helicopter.

[Objective] To provide a control device that can improve safety. [Means of Achieving the Goal] A control device 61 is provided in a vehicle system 2, which is logically divided into several subdivisions 64. The control device 61 comprises a semantic kernel (SK) 62, which controls communication access between two subdivisions 64 among several subdivisions 64 based on a static rule set; a rule set decision point (PDP) 66, which controls communication access between the two subdivisions 64 based on a dynamic rule set; and a rule set enforcement point (PEP) 68, which controls communication access between the two subdivisions 64 based on the control result of the PDP 66. If a predetermined condition is met, the PEP 68 forces the SK 62 to use the dynamic rule set instead of a part of the static rule set.

This invention provides a reinforcement learning-based intelligent adversarial system for game NPCs, belonging to the field of game adversarial systems. The system includes a data acquisition module that collects all raw interaction data from the target game engine and generates standardized trajectory data packets; a feature transformation module that uses a preset computational model for feature extraction and semantic encoding, and uses an adaptive segmentation algorithm to segment tactical fragments and mark target decision points; a causal path acquisition module that constructs a temporal causal discovery dataset, executes a preset temporal causal structure learning algorithm, generates a domain knowledge-enhanced causal graph, and acquires the target causal path; and a counterfactual inference module that proposes counterfactual intervention hypotheses, performs structured counterfactual inference using a preset inference method, and generates causal comparison training samples. This invention improves the information density of learning samples and the generalization ability of strategies, enriches game content, extends the game lifecycle, and directionally improves the intelligence level of NPCs without human intervention.

The application discloses a dynamic optimization method for identifying a key planning decision point of a comprehensive energy system, and relates to the field of power system dispatching and control. First, based on the physical composition of the comprehensive energy system, a planning and operation joint optimization model considering equipment operation constraints is established. Second, a certain equipment capacity scheme is taken as an input parameter, and the operation state in a long operation period is taken as evaluation, so that the value of the current planning scheme is calculated to form a value function model of the planning decision. Then, a rolling time domain optimization model of the comprehensive energy system is constructed, the value function model is called by the system in a rolling manner, the value in the current state is calculated, and a continuous value curve is obtained through linear interpolation and reverse recursion. The application converts the one-time complex investment problem into a multi-stage dynamic planning scheme of the comprehensive energy system through a non-decreasing incremental planning mode.

The application discloses a layered dynamic coordination-based on-board interference avoidance method and device. The method divides the interference avoidance task into interference avoidance in a multi-core processor and interference avoidance between multi-core processors. For the interference avoidance between multi-core processors, a high-layer scheduling controller performs interference avoidance according to the user density of adjacent spatial resource division units between the multi-core processors; for the interference avoidance in the multi-core processor, a media access control layer in the multi-core processor at a bottom layer is internally provided with a scheduling advance amount of each thread, so that different time points of adjacent threads on a pipeline correspond to the same time point of scheduling, resource allocation is performed at the same time point, a subsequent thread perceives a scheduling result of a previous thread, and interference avoidance is performed. The application reduces the calculation complexity of a single decision point, meets the real-time requirement of on-board scheduling, and significantly improves scheduling flexibility and interference response speed.

The application discloses an agent training method for embodied navigation decision understanding, aiming at solving the problem of insufficient generalization ability and decision-making ability of the embodied navigation agent in the prior art due to the dependence on single path imitation. The core of the application lies in two innovations: a) a novel training dataset generation process, which provides geodesic distance labeling based on the A* algorithm for all feasible candidate actions in the scene at each decision point of the agent, thereby constructing a dataset with "panoramic" supervision signals; b) an innovative interval perception hybrid reward function, which can dynamically assign reward signals according to the certainty of the current decision, providing strong guidance when the selection is clear, and providing detailed scoring to encourage exploration when the selection is ambiguous. The above method is applied to a two-stage training framework, which can guide the agent to change from "imitation path" to "understanding decision", and significantly improve the autonomous navigation performance of the agent in unknown environments.

A computer-implemented system and method for analyzing quality-event information through standardized representations, rule-based logic, and adaptive training. The system receives free-text quality-event descriptions and stores them in a historical database. A standardization module converts narratives into structured representations using domain terminology normalization and calibrated severity indicators. A rule-logic engine generates quality-rule constructs identifying primary, secondary, and systemic causes. A root-cause clustering module classifies events into related causal clusters. Based on identified clusters, the system generates training modules and control -measure templates for preventive or corrective actions. A simulation engine executes adaptive, branching scenarios presenting role-specific decision points and collecting performance metrics. A benchmarking engine computes comparative indicators across organizational units using anonymized data. An update module refines quality-rule constructs based on simulation performance and external system data. The system integrates with quality management, enterprise resource planning, and training platforms, enabling continuous organizational improvement through objective, data-driven analysis and dynamic, personalized training delivery.

Systems and methods include an interactive environment to develop different workflows in accordance with various entry points, actions, decisions, questions, answers, and / or the like in order to guide one or more users through a decision making process, which may be based on certain thresholds to assess one or more risks or other outcomes. A corpus of information may be used to direct a workflow through a series of questions, decision points, and then subsequent questions and / or workflows based, at least in part, on analysis of various thresholds or conditions, which may be defined by the workflow and / or within the corpus. In this manner, users may generate individual workflows for a given assessment and then apply the workflow to a variety of different end users.

This invention provides a machine translation method based on entropy-driven unsupervised reinforcement learning, belonging to the fields of natural language processing and machine translation. It aims to solve the problems of pattern collapse and consensus bias caused by the lack of reference translations in existing unsupervised machine translation reinforcement learning. The method includes: constructing the source sentence as an input sequence with a fine-tuning instruction format; during the autoregressive generation process, using word-level entropy as uncertainty feedback to dynamically adjust the sampling temperature and generate a diverse set of candidate translations; combining majority voting consensus and the model's intrinsic confidence to construct a confidence-weighted hybrid distribution, from which high-quality pseudo-labels are selected as supervision signals; in the policy optimization stage, entropy is used to identify high-uncertainty positions, and the advantage function is reweighted to focus on optimizing highly difficult semantic decision points. This invention can achieve the co-evolution of reward signals and model capabilities under unsupervised conditions, effectively avoiding the homogenization of generated results and significantly improving the quality of machine translation.

This invention discloses a job shop scheduling method based on heterogeneous graph neural networks and priority post-experience replay, relating to the field of dynamic job shop scheduling technology. By loading instances of the dynamic job shop scheduling problem and initializing deep reinforcement learning model parameters, state encoding, policy execution, environmental interaction, and experience storage operations are performed at decision points during training rounds. After each round, synthetic experience is generated through target relabeling, and a hybrid experience replay buffer containing both original and synthetic experience is constructed, with priorities assigned based on temporal difference errors. Experience samples are sampled according to priority, and network parameters are optimized using the PPO algorithm. The optimal model parameters are then loaded to perform real-time scheduling decisions for new instances. This invention improves sample utilization efficiency and policy convergence speed, exhibiting excellent scheduling optimization performance in both static and dynamic environments.

The invention relates to a vehicle alternative path generation method and device and computer equipment. A structured decision point network is constructed in a road space by utilizing precise road information provided by a high-precision map, map constraints such as lane lines and road structures are embedded into a decision point set, geometric and semantic boundaries are provided for path generation, the feasibility and constraint compliance of a path are improved, and when a next decision point of the path is connected, the path is more accurate to generate. And vehicle kinematics and dynamics constraints are comprehensively considered. A target point with a good obstacle avoidance effect and stable transverse change is preferentially selected through a decision point selection strategy, all obstacle information is projected into the decision points in a preprocessing stage before path searching, and each decision point contains the obstacle information. When collision detection is carried out, collision judgment can be completed by reading the structured information on the decision points, and the collision detection efficiency and the path planning response speed are greatly improved.

The invention discloses an intelligent conference record processing method, system and device and a medium, and the method specifically comprises the steps: building a causal atlas through an extraction event based on a primary text and a conference audio stream, marking decision points according to the causal atlas, and obtaining a logic chain analysis result; based on the primary text and the conference audio stream, analyzing emotional fluctuation of the speaker in combination with the acoustic features and the text to obtain an emotional voiceprint analysis result; based on a logic chain analysis result and an emotional voiceprint analysis result, adopting a large language model guided by a specific Prompt to generate an initial conference summary abstract attached with an original text sentence sequence label as a source tracing corner mark; and receiving user role configuration information, processing the initial conference summary abstract according to the user role configuration information, and generating a granularity-adjustable target conference summary abstract matched with the user role. The conference recording efficiency and quality are improved, and diversified requirements are met.

A control system detects an agricultural machine position to determine when the agricultural machine is approaching a decision point. The decision point is a point at which the agricultural machine can move forward along one of two or more predefined possible paths. The control system detects agronomic factors and vehicle-related parameters to decide on the path to take at the decision point. The control system then generates control signals to pursue the path decided upon.

The application provides an abnormal traffic analysis device based on a knowledge graph, and relates to the technical field of network security, which extracts instruction execution tracks capable of reflecting real execution logic, and uniformly maps instructions under different architectures into semantic fingerprint vectors, combines function labels, jump relationships and time sequence relationships to construct an instruction correlation graph, and expresses business semantics and control paths behind traffic in a graph mode; on this basis, the graphs of multiple executors are compared for consistency, and the abnormality degree is quantitatively evaluated in combination with security sensitivity, time sequence causal relationship, key path deviation and logic behavior baseline, to generate an abnormal trust score, and then the access control and executor disposal result are output by a strategy decision point, so that accurate identification, credible determination and linkage protection of abnormal traffic are realized.

A method for operating an autonomous vehicle may include generating a multitude of possible paths. The method may further include identifying one or more decision points along each of these possible paths. The method may also include determining a complexity score for each of the one or more decision points along each of the possible paths. The method may further include determining an optimal path, at least partially, based on the complexity score of each of the one or more decision points along each of the possible paths. Finally, the method may include adjusting the operation of the autonomous vehicle, at least partially, based on the optimal path.

This invention relates to the field of information system detection technology, specifically disclosing an artificial intelligence-based method for detecting anomalies in information system operation. The method includes: dividing the historical intelligent decision-making chain of an auctioned property into decision interval categories, statistically analyzing the proportion of anomalous decision points in each interval, calculating the confidence level anomaly probability, setting different detection window lengths, and then using the detection window length as a fixed unit to slide and extract the decision chain to be judged to generate a window to be judged. Continuous anomalies within the judgment window are identified, and new anomalous decision points are marked. The window to be reviewed is identified by combining the interval confidence level anomaly probability. Subsequently, the offline due diligence results chain is linked to compare and verify the validity of the anomalies, distinguishing between misjudgments and genuine anomalies. Finally, the initial detection window length is dynamically adjusted based on the time characteristics of the anomalous decision points to obtain the final detection window length, thus achieving dynamic adjustment of the detection window length and improving the accuracy of anomaly detection and risk control capabilities.

This invention innovatively proposes a core operating mechanism of "loosening restrictions initially and tightening them later, with soft constraints to prevent reverse deviation." Before the vehicle's actual cumulative mileage reaches the deviation detection trigger threshold, reverse road segments are eliminated solely through global directional soft constraints, without applying forced directional guidance, maximizing the flexibility of local path optimization. Once the vehicle's cumulative mileage to the decision point exceeds the preset threshold, global deviation detection and adaptive directional guidance correction are immediately initiated. Simultaneously, relying on a liquid neural network, online adaptive optimization of the congestion weight coefficient and the global directional guidance weight coefficient is achieved, dynamically adjusting the overall path cost function. This invention effectively avoids the drawbacks of excessively strong constraints in the early stages of traditional algorithms and completely solves the problems of reverse detours and local deadlocks caused by purely delayed trigger detection. It can balance local and global path optimization, exhibiting strong real-time performance and outstanding robustness, and can be applied to vehicle route navigation scenarios during evening rush hour.

This invention provides a system and method for implementing a multi-agent reinforcement learning-based decision system for material handling systems. [Solution] The method includes initializing a simulation environment that includes decision points and attributes for material dispatch of a material handling system, and which requests a multi-agent reinforcement learning-based decision system to make decisions for material dispatch at the decision points; initializing a reinforcement learning agent that represents the decision points; initializing domain expert heuristics for the decision points; and iteratively training the multi-agent reinforcement learning-based decision system using the domain expert heuristics initialized in the simulation environment.

The invention discloses a terrain decision point intelligent analysis method and system based on a neural operator, and a storage medium. The method comprises the following steps: preprocessing multi-source static terrain data, unifying coordinates and resolution, and generating derived indexes such as a road distance field and a reachable cost function; inputting the standardized multi-channel raster data into a Fourier neural operator model, and outputting at least one continuous utility field of visibility, safety and accessibility; candidate points are generated based on morphological features and road topology, and forbidden areas, minimum spacing and category mutual exclusion hard constraints are applied; taking the compliance candidate point as an action space, and adopting a sequence combination optimization strategy with an action mask to select no more than K decision points; according to the method, automation, compliance and global optimal selection of terrain decision points are achieved, the point selection accuracy and efficiency are improved, and the method is suitable for scenes such as situation simulation, deployment deduction and path organization and is high in practicability and adaptability.