Optimization of Aircraft Trajectory Design Based on Intelligent Learning Methods

This paper proposes an adaptive trajectory optimization algorithm based on deep reinforcement learning (DRL) to address the issues of low computational efficiency and poor environmental adaptability in aircraft trajectory optimization. The algorithm constructs a multi-layer neural network model and incorporates an improved policy gradient method to achieve real-time trajectory planning in complex environments. A dynamic reward mechanism and hierarchical decision structure are introduced, effectively enhancing the system’s adaptability to extreme weather and sudden disturbances. Large-scale testing and verification on a high-precision simulation platform demonstrate that the proposed algorithm outperforms traditional methods in terms of computational efficiency, optimization performance, and system stability, with significant engineering application value.