Artificial Intelligence–Assisted Aerodynamic Optimization of Toroidal Propellers for Enhanced Drone Performance

Abstract

The rapid advancement of unmanned aerial vehicles (UAVs) has increased the demand for efficient, low-noise, and aerodynamically optimized propulsion systems. Toroidal propellers, characterized by their ring-shaped blade geometry, have emerged as a promising alternative to conventional propellers due to their potential for vortex reduction and noise suppression. This study integrates theoretical aerodynamic modeling with Artificial Intelligence (AI)-based optimization techniques to evaluate the performance and feasibility of toroidal propellers for drone applications. Mathematical thrust models for both traditional and toroidal propellers were developed and implemented using MATLAB simulations. Machine learning algorithms were applied to predict thrust performance across varying diameters and operating conditions. The AI-assisted analysis demonstrated improved thrust characteristics and reduced turbulence in toroidal designs. The results indicate that integrating AI with aerodynamic modeling enhances performance prediction accuracy and supports design optimization. This research highlights the transformative potential of toroidal propellers combined with computational intelligence for next-generation drone systems.