一种基于双重深度Q学习的航空器实时协同航迹规划方法
By employing a trajectory planning method based on dual deep Q-learning, an intelligent agent was designed and trained to achieve collaborative trajectory planning for multiple aircraft under adverse weather conditions. This solves the real-time and security issues of trajectory planning in existing technologies and improves the automation and intelligence level of aircraft operations.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
- Filing Date
- 2023-12-22
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to achieve real-time, safe, and efficient trajectory planning in collaborative trajectory planning for multiple aircraft, especially lacking universally applicable tactical trajectory planning algorithms under adverse weather conditions.
A real-time collaborative trajectory planning method for aircraft based on dual deep Q-learning is adopted. By designing a trajectory planning agent, training the agent using a dual deep Q-learning algorithm with a commenter network, and combining a reward function and a state transition function, a conflict-free and refined four-dimensional trajectory planning for aircraft is achieved.
It improves the adaptability and real-time performance of trajectory planning, reduces the workload of air traffic controllers, and enhances the automation and intelligence of trajectory planning, especially maintaining efficient operation under random and dynamic weather conditions.
Smart Images
Figure CN117824649B_ABST