UAV Formation Flight Using Non-Uniform Vector Field and Fuzzy Self-Tuning PD-Control
Tagir Z. Muslimov, Rustem Anvarovich Munasypov · 2018 International Russian Automation Conference (RusAutoCon) · 2018
The paper presents a new three-dimensional formation flight control algorithm for fixed-wing unmanned aerial vehicles (UAVs) based on non-uniform guidance vector field. Consensus-based model of living organisms' motor neurons network was used for generalized law of UAVs interaction. A non-uniform vector field is constructed for obtaining desired velocities and course angles in order to take into account the nonlinear structure of the “low-level autopilot/UAV” system. The stability analysis with Lyapunov direct method shows that relative distance errors for each UAV approach zero asymptotically. Then, model reference self-tuning fuzzy PD-control was used in order to ensure the desired quality of transient trajectories in the case when one or more vehicles' dynamics is significantly different from those vehicles' dynamics for which the coefficients in the group control algorithm were selected. The adaptive loop stability is provided with fuzzy Lyapunov synthesis. The overall control system performance was tested in MATLAB/Simulink using full dynamic of the low-level autopilot/UAV nonlinear models.