一种跨介质飞行器飞行状态智能识别方法
By introducing time axis constraints into the flight status identification of cross-medium aircraft, and employing shallow neural networks and feature attribute screening methods, the problem of poor identification robustness in existing technologies is solved, achieving accurate and stable identification of hypersonic cross-medium aircraft, and the network design is lightweight and energy-efficient.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NORTHWESTERN POLYTECHNICAL UNIV
- Filing Date
- 2024-03-25
- Publication Date
- 2026-07-17
AI Technical Summary
Existing cross-medium aircraft flight status identification methods are not robust enough in supersonic and high-speed aircraft. Data-driven methods ignore the relationship on the time axis, resulting in unsatisfactory identification results.
By constructing time-axis-based constraints, a shallow neural network is used to identify the flight status of cross-media aircraft. Features are selected using the information gain and information gain ratio of feature attributes, and a neural network with convolutional layers, max pooling layers, and fully connected layers is built for identification.
It achieves accurate identification of hypersonic, highly maneuverable, and rapidly changing cross-medium aircraft, ensuring the accuracy and stability of identification. At the same time, the designed network is lightweight, energy-efficient, and highly effective.
Smart Images
Figure CN118520380B_ABST