一种基于强化学习的无人机自主目标跟踪控制方法、系统及设备
By constructing an autonomous tracking reinforcement learning task model and designing the TD3 reinforcement learning training method, combined with teacher-guided annealing training mechanism and multidimensional reward function, the problem of high-precision tracking of UAVs in complex environments was solved, achieving smooth and stable tracking of UAVs and avoiding the risk of loss of control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CALCULATION AERODYNAMICS INST CHINA AERODYNAMICS RES & DEV CENT
- Filing Date
- 2026-04-23
- Publication Date
- 2026-07-17
AI Technical Summary
Existing reinforcement learning-based UAV control methods struggle to achieve high-precision autonomous tracking in complex environments and are prone to issues such as unsmooth trajectories and control divergence. Furthermore, existing reward function designs fail to adequately consider the relative geometric relationship between the UAV and the target.
An autonomous tracking reinforcement learning task model was constructed, the TD3 reinforcement learning training method was designed, a teacher-guided annealing training mechanism was introduced, and flight control constraints and multi-dimensional reward functions were designed. Through joint optimization of the policy network and the dual-value network, a target UAV tracking and control model was established.
It achieves high-precision autonomous tracking of drones in complex environments, avoiding uncontrolled behaviors such as overspeeding and rapid acceleration, and ensuring safety and smooth operation.
Smart Images
Figure CN122086092B_ABST