Adaptive Multi-Layered Trajectory Planning for Autonomous Aerial Vehicles

Autonomous Aerial Vehicles (AAVs) play a pivotal role in modern applications such as disaster response, surveillance, logistics, and environmental monitoring, offering high efficiency, adaptability, and safety in dynamic and unpredictable environments. A critical component of AAV functionality is robust trajectory planning, which ensures accurate path prediction, effective obstacle avoidance, and energy-efficient navigation. This paper proposes a hybrid trajectory planning framework that integrates global path optimization, local obstacle avoidance, and controlled descent to improve landing precision and overall mission efficiency. To further enhance adaptability in real-world conditions, the framework incorporates a reinforcement learning module that dynamically fine-tunes trajectory adjustments in response to external disturbances. Experimental results demonstrate that the proposed multilayered system significantly improves AAV performance in complex and volatile operational scenarios.