4D Trajectory Regulation via Knowledge Graph Similarity Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing air traffic control system struggles to meet increasing flight demands, leading to congestion and delays, and relies heavily on experienced air traffic controllers for real-time 4-dimensional trajectory management, which is risky and inefficient.
Innovation Solution
A 4-dimensional trajectory regulatory decision-making method that converts air traffic systems into graph data structures using knowledge graphs, processing node vectors to match similar historical scenes and retrieve decision-making instructions for current flight operations, reducing reliance on human experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If air traffic control system evolves towards 4-dimensional trajectory based operation, then flight management capability and safety are improved, but system complexity and difficulty of control increase
Solution Approach 1:
The patent segments the complex air traffic control system into multiple knowledge graphs (flight knowledge graph, airspace knowledge graph, airport knowledge graph, etc.), each handling specific aspects of flight management. This segmentation reduces the complexity of individual components while maintaining overall system capability for 4-dimensional trajectory management.
Solution Approach 2:
The patent introduces a regulatory decision-making system as an intermediary between air traffic controllers and the complex 4-dimensional trajectory management requirements. This system uses knowledge graphs and similarity matching to provide automated decision support, reducing the burden on controllers while ensuring safety and reliability.
2Measurement precision
If air traffic control relies on experienced controllers for real-time decision making, then regulatory accuracy is improved, but operational risk and dependency on human experience increase
Solution Approach 1:
The patent copies expert decision-making knowledge into structured knowledge graphs that can be systematically stored and retrieved. By encoding regulatory rules, flight procedures, and decision logic into the knowledge graph structure, the system reproduces expert-level accuracy without relying on individual human controllers' experience, thereby improving operational reliability.
Solution Approach 2:
The system implements feedback mechanisms where historical flight data and regulatory outcomes are continuously incorporated into the knowledge graph. This allows the system to learn from past decisions and improve regulatory accuracy over time, while providing consistent, reliable decision support independent of individual controller variability.
3Speed
If real-time knowledge graph processing is implemented, then decision-making speed is improved, but computational complexity increases
Solution Approach 1:
The patent performs preliminary processing by pre-constructing knowledge graphs from historical data and regulatory rules before real-time operation. During actual flight management, the system queries these pre-processed knowledge graphs rather than processing raw data in real-time, significantly reducing computational complexity while maintaining fast decision-making speed.
Solution Approach 2:
The patent transforms complex multi-dimensional flight data into structured knowledge graph representations with defined schemas and relationships. This dimensional transformation organizes unstructured data into query-optimized formats, reducing computational complexity for real-time operations while preserving all necessary information for fast decision-making.
4Measurement precision
If historical flight data is extensively utilized for similarity matching, then regulatory accuracy is improved, but data processing time increases
Solution Approach 1:
The patent pre-processes historical flight data during off-peak periods to build the knowledge graph structure and index similar flight scenarios. When real-time regulatory decisions are needed, the system queries this pre-organized structure rather than analyzing raw historical data, significantly reducing data processing time while maintaining high regulatory accuracy through similarity matching.
Data Source
AI summary
The present application provides a 4-dimensional trajectory regulatory decision-making method for air traffic. The method includes: acquiring a node vector set of a target flight to obtain a target node vector set; acquiring the node vector set of all historical flights simultaneously to obtain a historical node vector set; acquiring a similarity between the target node vector set and the node vector set of each historical flight in the historical node vector set to obtain a plurality of similarity data; acquiring a similarity data which is greater than a preset similarity data from plurality of similarity data to obtain a target similarity data; acquiring a historical flight to which the target similarity data belongs to obtain a target historical flight; acquiring a historical decision-making instruction of the target historical flight to obtain a target decision-making instruction.