Optimization of PID Controllers for UAVs Using SPSA Algorithm for Enhanced Flight Performance

This paper presents the design and optimization of a longitudinal control system for a fixed-wing Unmanned Aerial Vehicle (UAV). The study focuses on the development of a state-space model based on the UAV's aerodynamic parameters. The system matrices (A and B) are derived from the vehicle's physical properties and aerodynamic coefficients, allowing for an accurate representation of the UAV's response to control inputs. A PID controller was used to regulate the pitch angle, and its parameters are optimized using the Simultaneous Perturbation Stochastic Approximation (SPSA) method. This optimization approach proves to be effective in fine-tuning the control gains by minimizing the error in the pitch response under realistic flight conditions. The study emphasizes the robustness of SPSA, particularly in high-dimensional and noisy environments.The results demonstrate the improved autonomous performance of the UAV, with the PID controller successfully achieving the desired pitch angle control.