融合引导点奖励机制的舰载机调运路径深度强化学习规划方法
By combining deep reinforcement learning and Bézier curve-guided path, the problems of real-time monitoring and path smoothness in carrier-based aircraft scheduling were solved, enabling efficient, safe, and smooth operation of carrier-based aircraft in complex obstacle environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- DALIAN UNIV OF TECH
- Filing Date
- 2026-05-13
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to monitor the motion status of carrier-based aircraft in real time, are prone to collisions, have long computation times, and lack path smoothness, especially in complex obstacle environments where computation time increases.
A deep reinforcement learning approach is adopted to generate guide point paths through third-order Bézier curves. Combined with a dense reward function with dynamic weight adjustment, the policy-value network is trained using the PPO algorithm to achieve end-to-end planning of carrier-based aircraft paths, and to generate smooth paths that meet kinematic and obstacle avoidance constraints.
It enables efficient, safe, and smooth planning of carrier-based aircraft transport paths, and can generate efficient, safe, and smooth transport trajectories in complex deck environments, thereby improving the real-time performance and computational efficiency of carrier-based aircraft scheduling.
Smart Images
Figure CN122175123B_ABST