Complex dynamic scene-oriented degree space adaptive flexible decision regulation and control system
By designing a spatially adaptive and flexible decision-making and control system in complex dynamic scenarios, the problem that traditional systems cannot effectively integrate multi-dimensional and multi-space data is solved, and efficient and accurate decision-making and control are achieved.
Patent Information
- Application Number
- CN202510584088.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-07
- Publication Date
- 2025-06-24
AI Technical Summary
When facing complex dynamic scenarios, traditional decision-making systems cannot effectively integrate multi-dimensional and multi-space multi-modal data, resulting in a lack of comprehensiveness and accuracy in decision-making results, low computing efficiency and low degree of automation.
A dimensional space adaptive flexible decision-making and regulation system for complex dynamic scenarios is designed. Through multi-space data input module, multi-space mapping module, decision fusion module and scheduling control module, the fusion of multi-space data and adaptive regulation of decision-making are realized.
It significantly improves the accuracy, flexibility and computing efficiency of decision-making, and can quickly respond in complex dynamic environments and achieve efficient resource allocation and task scheduling.
Smart Images
Figure CN120197134A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of information technology, and particularly to a degree-space adaptive flexible decision-making and control system for complex dynamic scenarios. Background Art
[0002] In the decision-making and control tasks of complex dynamic scenarios, traditional decision-making systems often rely on a single data source or a fixed model, lacking the ability to effectively process multi-dimensional and multi-space data. Such systems usually make decisions based on simple linear relationships or predefined rules. Although they can provide reasonable decision support in relatively simple environments, their response speed and decision accuracy often cannot meet the actual needs when facing rapidly changing environments or involving multi-dimensional and multi-modal data.
[0003] Specifically, the traditional decision-making system has the following main problems: Single data source limitation: Traditional systems usually can only process information from a single data source and cannot effectively fuse multi-modal data from different spaces (such as Euclidean space, hyperbolic space, hypersphere space, etc.), resulting in the lack of comprehensiveness and accuracy of decision-making results.
[0004] Lack of flexibility: Traditional systems often rely on fixed decision-making models and rules and cannot adjust decision-making strategies in real time according to environmental changes, resulting in the inability to make a rapid response when facing complex dynamic scenarios.
[0005] Low computational efficiency: When processing large-scale or complex data, traditional systems usually require a long computational time and cannot meet the needs of real-time decision-making.
[0006] Low degree of automation: Traditional systems often require a large amount of manual intervention to make effective decisions and resource scheduling, increasing the operation and maintenance costs and time costs.
[0007] To solve the above problems, the present invention proposes a degree-space adaptive flexible decision-making and control system for complex dynamic scenarios. The system realizes efficient and accurate decision-making and control in complex dynamic scenarios by introducing multi-space fusion, adaptive adjustment algorithms, and real-time feedback mechanisms. Summary of the Invention
[0008] The object of the present invention is to provide a degree-space adaptive flexible decision-making and control system for complex dynamic scenarios, which solves the problem that traditional systems in the prior art usually can only process information from a single data source and cannot effectively fuse multi-modal data from different spaces, resulting in the lack of comprehensiveness and accuracy of decision-making results.
[0009] To achieve the above object, the present invention adopts the following technical solutions: A degree-space adaptive flexible decision-making and control system for complex dynamic scenarios, including a multi-space data input module, a multi-space mapping module, a decision fusion module, and a scheduling control module; the multi-space data input module is used to collect data from different decision spaces; the multi-space mapping module is used to map the collected data to different spaces for feature extraction; the decision fusion module is used to fuse the feature vectors extracted from different spaces and generate a decision result; the scheduling control module is used to perform real-time control and scheduling of the system according to the decision result.
[0010] Preferably, the data acquisition unit is used to collect data from different sources such as sensors, user inputs, and environmental monitoring; the data preprocessing unit is used to denoise, convert the format, and perform preliminary processing on the raw data to ensure that the data can be effectively converted into a format suitable for subsequent analysis.
[0011] Preferably, the multi-space mapping module includes an Euclidean space mapping unit, a hyperbolic space mapping unit, and a hypersphere space mapping unit; the Euclidean space mapping unit is used to map the processed raw data to the Euclidean space for feature extraction; the hyperbolic space mapping unit is used to map the processed raw data to the hyperbolic space for feature extraction; the hypersphere space mapping unit is used to map the processed raw data to the hypersphere space for feature extraction.
[0012] Preferably, the decision fusion module includes a weighting processing unit, an optimal transport fusion unit, and a scheduling control signal generation unit; the weighting processing unit is used to perform weighting processing on the feature vectors in different spaces; the optimal transport fusion unit is used to fuse the weighted feature vectors through the optimal transport algorithm and generate a preliminary decision result; the scheduling control signal generation unit is used to convert the preliminary decision result into a scheduling control signal and transmit it to the scheduling control module.
[0013] Preferably, the feedback mechanism module includes a state monitoring unit and a feedback signal generation unit; the state monitoring unit is used to monitor the execution state of the system in real time; the feedback signal generation unit is used to generate a feedback signal according to the execution state of the system and transmit the feedback signal to the decision fusion module to adjust the decision-making strategy.
[0014] Preferably, the adaptive adjustment mechanism includes a critical path discovery unit and an adaptive optimization adjustment unit; the critical path discovery unit is used to identify the critical path according to the feedback information of the current environment; the adaptive optimization adjustment unit is used to dynamically adjust the weight coefficients of each space through a multi-space adaptive algorithm to optimize the decision-making path.
[0015] Preferably, the multi-space adaptive algorithm includes a reinforcement learning algorithm or a genetic algorithm; the reinforcement learning algorithm is used to optimize the weight coefficients by continuous trial and error and learning; the genetic algorithm is used to optimize the weight coefficients by simulating natural selection and genetic mechanisms.
[0016] Preferably, it further includes a user interface module, which is used to interact with the user, receive the input instructions of the user and display the decision results and execution status of the system.
[0017] The present invention has the following beneficial effects: The degree-space adaptive flexible decision-making and regulation system of the present invention for complex dynamic scenarios significantly improves the accuracy, flexibility and computational efficiency of decision-making through technical means such as multi-space fusion, adaptive adjustment and efficient calculation, and provides an efficient solution for dealing with resource allocation and task scheduling problems in complex dynamic environments. Brief Description of the Drawings
[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.
[0019] Figure 1 It is a schematic diagram of the core framework of the present invention. Detailed Embodiments
[0020] In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Embodiment
[0021] Please refer to Figure 1 As shown, the degree-space adaptive flexible decision-making and regulation system of the present embodiment for complex dynamic scenarios includes a multi-space data input module, a multi-space mapping module, a decision fusion module and a scheduling control module; the multi-space data input module is used to collect data from different decision spaces; the multi-space mapping module is used to map the collected data to different spaces for feature extraction; the decision fusion module is used to fuse the feature vectors extracted from different spaces and generate decision results; the scheduling control module is used to perform real-time control and scheduling of the system according to the decision results.
[0022] Multi-space data input stage: The system collects data from different decision spaces (such as Euclidean space, hyperbolic space, hypersphere space) and converts the data into a unified feature representation through different space mapping methods.
[0023] The specific steps include: obtaining the data representation in Euclidean space, obtaining the data representation in hyperbolic space, obtaining the data representation in hypersphere space, and fusing the feature vectors of the three spaces.
[0024] Multi-space decision regulation stage: Based on the fused multi-space feature vectors, the system generates the final decision result by constructing a decision model and an adaptive optimization algorithm.
[0025] The specific steps include: constructing a decision model, discovering and adaptively optimizing the critical path, and the flexibility and accuracy of multi-space regulation.
[0026] Multi-space fusion: By fusing the data features from different spaces, the system can capture more-dimensional decision information, improving the accuracy and comprehensiveness of decision-making.
[0027] Adaptive adjustment: The system can dynamically adjust the decision according to environmental changes and real-time feedback, enhancing the flexibility and adaptive ability of the system.
[0028] Efficient computation: Through optimized algorithms and distributed computing architectures, the system can significantly improve efficiency when processing large-scale dynamic data.
[0029] Automated regulation: The system has a strong ability of automated regulation and can achieve efficient decision-making and resource scheduling without frequent manual intervention.
[0030] Multi-space data input stage: In this stage, the system collects data from different decision spaces and converts the data into a unified feature representation through different space mapping methods. Specifically, this stage includes three main spaces: Euclidean space, hyperbolic space, and hypersphere space. Each space represents a different data distribution method. Through the data mapping of these spaces, the system can capture the relationships between different spaces and provide a basis for subsequent decision regulation.
[0031] Step 1: Obtaining the data representation in Euclidean space In Euclidean space, data points have regular flat geometric properties, so their features can be calculated through standard distance metrics (such as Euclidean distance). Assume the original data collected from Euclidean space is , where is a data point in Euclidean space. The system obtains its feature vector by normalizing the data (such as normalization or denoising).
[0032] Step 2: Obtaining the data representation in hyperbolic space Hyperbolic space has negative curvature, and data exhibits different geometric features in this space. The distance metric in hyperbolic space is different from that in Euclidean space, so the way of data mapping is also different. In this step, it is assumed that the data collected from the hyperbolic space is , and by using a hyperbolic geometry model (such as the Poincaré model) for mapping, the eigenvector in the hyperbolic space is obtained . Embodiment
[0033] Please refer to Figure 1 As shown, for the degree space adaptive flexible decision regulation system facing complex dynamic scenarios in this embodiment, the data acquisition unit is used to collect data from different sources such as sensors, user inputs, and environmental monitoring; the data preprocessing unit is used to denoise, convert the format, and perform preliminary processing on the raw data to ensure that the data can be effectively converted into a format suitable for subsequent analysis.
[0034] The multi-space mapping module includes an Euclidean space mapping unit, a hyperbolic space mapping unit, and a hypersphere space mapping unit; the Euclidean space mapping unit is used to map the processed raw data to the Euclidean space for feature extraction; the hyperbolic space mapping unit is used to map the processed raw data to the hyperbolic space for feature extraction; the hypersphere space mapping unit is used to map the processed raw data to the hypersphere space for feature extraction.
[0035] The decision fusion module includes a weighting processing unit, an optimal transport fusion unit, and a scheduling control signal generation unit; the weighting processing unit is used to perform weighting processing on the eigenvectors in different spaces; the optimal transport fusion unit is used to fuse the weighted eigenvectors through the optimal transport algorithm and generate a preliminary decision result; the scheduling control signal generation unit is used to convert the preliminary decision result into a scheduling control signal and transmit it to the scheduling control module.
[0036] It also includes a feedback mechanism module, and the feedback mechanism module includes a state monitoring unit and a feedback signal generation unit; the state monitoring unit is used to monitor the execution state of the system in real time; the feedback signal generation unit is used to generate a feedback signal according to the execution state of the system and transmit the feedback signal to the decision fusion module for adjusting the decision-making strategy.
[0037] Improving Decision Accuracy and Adaptability: By combining the multi-modal data features of Euclidean space, hyperbolic space, and hypersphere space, the present invention effectively maps and fuses data using the geometric properties of different spaces, thereby being able to capture more dimensional decision-making information and greatly improving the accuracy of decision-making. Traditional methods usually rely on a single data source or spatial model and lack multi-angle and multi-level analysis of data. Through the multi-space adaptive decision-making model, the present invention can select the most suitable space for analysis according to different properties of the environment (such as spatial distribution, task complexity, etc.), thus obtaining more accurate decision-making results.
[0038] Enhancing System Flexibility and Adaptive Capability: A core advantage of the present invention is that the system can dynamically adjust decisions according to environmental changes and real-time feedback. By introducing an adaptive regulation mechanism, the system can quickly adjust decision-making strategies when task requirements change, ensuring high-efficiency and precise regulation capabilities in dynamically changing complex scenarios.
[0039] Different from traditional static decision-making systems, this system can not only make preliminary decisions based on external conditions but also adjust strategies according to real-time feedback during the execution process, avoiding the inefficient response of traditional systems in the face of emergencies.
[0040] Step 3: Obtain the data representation of the hypersphere space The hypersphere space has positive curvature and is usually used to represent complex data structures. In this space, the mapping of data needs to be transformed through the angles and radii on the sphere. Assume the data collected from the hypersphere space is , and through the mapping method of the hypersphere space, it is transformed into the corresponding feature vector .
[0041] Step 4: Fuse the feature vectors of the three spaces The feature vectors , , and obtained from Euclidean space, hyperbolic space, and hypersphere space respectively are fused to obtain a unified multi-space feature vector , which will be used as the input in the decision regulation process.
[0042] Multi-space Decision Regulation Stage: In this stage, the system generates the final decision result according to the feature vector F\ fused from data from different spaces. The goal of this stage is to achieve decision regulation through the feature information of multiple spaces, enabling the system to flexibly adapt to different dynamic scenarios.
[0043] Step 5: Construct a decision-making model Based on the multi - space feature vector F, the system calculates the decision result by constructing a multi - space decision model (such as a weighted decision model, an optimization - based regulation model, etc.). The decision model weights according to the weight coefficients of different space features to generate the final decision vector D. Suppose the weights of the space features are , , and , then the decision vector can be expressed as: Embodiment
[0044] Please refer to Figure 1 As shown, for the degree - space adaptive flexible decision - regulation system for complex dynamic scenarios in this embodiment, the adaptive adjustment mechanism includes a critical - path discovery unit and an adaptive optimization adjustment unit; the critical - path discovery unit is used to identify the critical path according to the feedback information of the current environment; the adaptive optimization adjustment unit is used to dynamically adjust the weight coefficients of each space through a multi - space adaptive algorithm, so as to optimize the decision path.
[0045] The multi - space adaptive algorithm includes a reinforcement learning algorithm or a genetic algorithm; the reinforcement learning algorithm is used to optimize the weight coefficients by continuous trial - and - error and learning; the genetic algorithm is used to optimize the weight coefficients by simulating natural selection and genetic mechanisms.
[0046] It also includes a user interface module, which is used to interact with users, receive user input instructions and display the decision results and execution status of the system.
[0047] Efficient computing and processing capabilities: To meet the processing requirements of large - scale data, the present invention has been optimized in terms of computing efficiency. Through the combination of parallel computing, distributed computing frameworks, and approximate optimal transport algorithms, the system can significantly improve efficiency when processing large - scale dynamic data. In traditional decision - making systems, processing complex data often leads to long computing times and high computing costs, while the present invention enables the decision - making process to be completed at a higher speed and with lower resource consumption through efficient optimization algorithms and distributed computing architectures, adapting to the high - efficiency processing requirements in the big - data environment.
[0048] Requiring little manual intervention and having strong automation regulation capabilities: The present invention also has strong automation regulation capabilities. Through real - time feedback mechanisms and optimization algorithms, the system can autonomously adjust decision - making strategies and complete resource scheduling. This automation capability enables the system to achieve efficient decision - making and resource scheduling without frequent manual intervention in complex environments, significantly improving the operation and maintenance efficiency of the system.
[0049] Step Six: Critical - Path Discovery and Adaptive Optimization Adjustment Based on the feedback information of the current environment, the system first identifies the critical path and analyzes which decision paths are crucial for the overall performance of the system. The system dynamically adjusts the weight coefficients of each space through a multi-space adaptive algorithm , , and thus optimizing the decision paths. These paths (such as the critical path , etc.) can be identified through a graph structure or a multi-space fusion mechanism and weighted and adjusted according to their importance. To achieve this adaptive adjustment, the system adopts feedback-based optimization algorithms (such as reinforcement learning, genetic algorithms, etc.), and by continuously updating the weight coefficients, ensures that decisions can flexibly meet various requirements in a changing environment.
[0050] By automatically discovering the critical path and adjusting the weights, the system can accurately identify the decision paths with the most serious impact in the system, thereby performing resource allocation, task adjustment, and policy optimization. In this way, the decisions of the system are not only more efficient but also can achieve real-time response to environmental changes, avoiding the lag in traditional methods.
[0051] Step Seven: Flexibility and Precision of Multi-Space Regulation The final decision D is applied to specific regulation tasks for operations such as resource scheduling and task allocation. The system can simultaneously consider the characteristic information of multiple spaces (such as Euclidean space, hyperbolic space, hypersphere space) to ensure precise control and scheduling in a complex dynamic environment. This regulation process relies on real-time optimization in different decision spaces, further improving the flexibility and precision of the decision-making process.
[0052] By discovering and optimizing the critical path, the system can adjust the decision-making strategy according to real-time feedback. Whether it is for the optimization of network routing or applications in other dynamic scenarios, the system can adaptively adjust decisions according to environmental changes to ensure the efficient execution of the entire process.
[0053] Through the above method, the system can comprehensively consider the characteristic information of Euclidean space, hyperbolic space, and hypersphere space to achieve efficient and flexible decision regulation in complex dynamic scenarios, improving the decision precision and the ability of resource allocation.
[0054] The above has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.
Claims
1. A degree-space adaptive flexible decision-making and control system for complex dynamic scenes, characterized by: include: Multi-space data input module, multi-space mapping module, decision fusion module and scheduling control module; The multi-space data input module is used to collect data from different decision spaces; The multi-space mapping module is used to map the collected data to different spaces for feature extraction; the decision fusion module is used to fuse the feature vectors extracted from different spaces and generate a decision result; The scheduling control module is used to perform real-time control and scheduling of the system according to the decision results.
2. The degree space adaptive flexible decision-making and control system for complex dynamic scenes according to claim 1 is characterized in that: The data acquisition unit is used to collect data from different sources such as sensors, user input and environmental monitoring; the data preprocessing unit is used to remove noise, convert the format and perform preliminary processing on the raw data to ensure that the data can be effectively converted into a format suitable for subsequent analysis.
3. The degree space adaptive flexible decision-making and control system for complex dynamic scenes according to claim 2 is characterized in that: The multi-space mapping module includes a Euclidean space mapping unit, a hyperbolic space mapping unit and a hypersphere space mapping unit; the Euclidean space mapping unit is used to map the processed original data to the Euclidean space for feature extraction; the hyperbolic space mapping unit is used to map the processed original data to the hyperbolic space for feature extraction; the hypersphere space mapping unit is used to map the processed original data to the hypersphere space for feature extraction.
4. The degree space adaptive flexible decision-making and control system for complex dynamic scenes according to claim 3 is characterized in that: The decision fusion module includes a weighted processing unit, an optimal transmission fusion unit and a scheduling control signal generating unit; the weighted processing unit is used to perform weighted processing on feature vectors of different spaces; the optimal transmission fusion unit is used to fuse the weighted feature vectors through an optimal transmission algorithm and generate a preliminary decision result; the scheduling control signal generating unit is used to convert the preliminary decision result into a scheduling control signal and pass it to the scheduling control module.
5. The degree space adaptive flexible decision-making and control system for complex dynamic scenes according to claim 4 is characterized in that: It also includes a feedback mechanism module, which includes a state monitoring unit and a feedback signal generating unit; the state monitoring unit is used to monitor the execution state of the system in real time; the feedback signal generating unit is used to generate a feedback signal according to the execution state of the system, and pass the feedback signal to the decision fusion module to adjust the decision strategy.
6. The degree space adaptive flexible decision-making and control system for complex dynamic scenes according to claim 5 is characterized in that: The adaptive adjustment mechanism includes a critical path discovery unit and an adaptive optimization adjustment unit; The critical path discovery unit is used to identify the critical path according to the feedback information of the current environment; the adaptive optimization adjustment unit is used to dynamically adjust the weight coefficients of each space through a multi-space adaptive algorithm, so as to optimize the decision path.
7. The degree space adaptive flexible decision-making and control system for complex dynamic scenes according to claim 6 is characterized in that: The multi-space adaptive algorithm includes a reinforcement learning algorithm or a genetic algorithm; the reinforcement learning algorithm is used to optimize the weight coefficient through continuous trial and error and learning; the genetic algorithm is used to optimize the weight coefficient by simulating natural selection and genetic mechanisms.
8. The degree space adaptive flexible decision-making and control system for complex dynamic scenes according to claim 7 is characterized in that: It also includes a user interface module, which is used to interact with the user, receive the user's input instructions and display the system's decision results and execution status.