This paper introduces a novel framework for optimizing unmanned aerial vehicle (UAV) trajectories in 6G cellular systems by integrating enhanced continual transfer learning within the O-RAN architecture. The system utilizes a library of pre-trained models and a selection mechanism to minimize adaptation time when operating in dynamic environments.
- Maintains a library of pre-trained models and employs a model selection mechanism to identify relevant environments for knowledge transfer.
- Uses a fallback model with continuous refinements to ensure baseline performance when no sufficiently similar model is available.
- Leverages real-world city maps and ray tracing techniques to enhance learning reliability and trajectory planning.
- Reduces convergence time by 44% to 56% compared to retraining from scratch, and up to 40% compared to traditional transfer learning without model selection.