This work focuses on advancing the integration of autonomous aircraft into civilian airspace, a key component of NASA’s Urban Air Mobility (UAM) initiative. Previously, I developed a cooperative planning and control framework that ensures collision avoidance and energy-efficient autonomous flight using optimal control and formal safety guarantees. This framework provides mathematically rigorous methods for generating safe trajectories in constrained and dynamic environments. Building on this foundation, I am now investigating how neural networks can enhance trajectory generation by improving adaptability to dynamic airspace conditions. Traditional optimal control methods, while effective, often face computational challenges when applied in real-time. By integrating data-driven techniques, I aim to develop a framework that enables computationally efficient trajectory generation for autonomous aircraft in complex environments.