Three-dimensional trajectory design model of UAVs based on resource scheduling optimization algorithm

Unmanned Aerial Vehicles utilize their flexibility in flight and low-cost advantages in various fields to improve operational efficiency and coverage. However, during flight, environmental factors can cause trajectory deviation and reduce operational efficiency. This study applies the Particle Swarm Optimization algorithm as a resource scheduling optimization algorithm and proposes a three-dimensional trajectory design model for Unmanned Aerial Vehicles. The model combines the Particle Swarm Optimization algorithm and the Genetic Algorithm for resource scheduling and optimization. In addition, it enhances environmental perception by integrating fast independent component analysis and deep reinforcement learning algorithms to achieve precise three-dimensional trajectory design. Statistical results from the experimental verification of the proposed model show a recall rate of 94.79%, a precision rate of 92.72%, a safe distance rate of 99.52%, and a designed trajectory path length of 780.45 m. This indicates that the proposed model has strong capabilities in resource allocation, quick obstacle detection, and flight path adjustment. This will greatly support the technological development of Unmanned Aerial Vehicles.