Neural control of a Quadrotor: A state-observer based approach

CLAUDIO ROSALES; RICARDO CARELLI; CARLOS SORIA; FRANCISCO ROSSOMANDO

Dallas

Congreso; 2018 International Conference on Unmanned Aircraft Systems (ICUAS); 2018

IEEE

This paper proposes a neuro-adaptive controllerwith supervisory control for UAVs. A state observer of thecontrol errors is defined because the vehicle could be affectedby measurement noise and external disturbances. In this study,the control errors are estimated by a state neural observer fromwhich the neural adaptive system generates a control action.Also, a supervisor term is defined in order to guarantee thesystem stability. The neuro-adaptive controller parameters areupdated on-line by the adjustment laws that are a function ofthe control errors calculated by the observer. These adjustmentlaws were obtained through the Lyapunov stability criterion.To summarize, simulation results demonstrate the good performanceof the proposed technique for trajectory tracking tasksin navigation environments affected by measuring noise andexternal disturbances.