Adaptive trajectory tracking control for quadrotors with disturbances by using generalized regression neural networks

In this document, the development and experimental validation of a nonlinear controller with an adaptive disturbance compensation system applied on quadrotor are presented. The introduced scheme relies generalized regression neural network (GRNN). proposed has structure consisting inner control loop inaccessible to user (i.e., embedded controller) outer which generates commands for loop. GRNN is in approach lies aptitude estimate disturbances unmodeled dynamic effects without requiring accurate knowledge parameters. adaptation laws deduced from rigorous convergence analysis ensuring asymptotic trajectory tracking. implemented QBall 2 quadrotor. Comparisons respect PD-based control, model regressor-based scheme, neural-network carried out. results validate functionality novel show performance improvement since smaller tracking error values produced.