Purpose
The purpose of this paper is to design a controller for a quadrotor only by using input–output data without a need for the system model.
Design/methodology/approach
Tracking control for the quadrotor is considered by using unfalsified control, which is one of the most recent strategies of robust adaptive control. The main assumption in unfalsified control design is that there is no access to the system model. Also, ideal path tracking and controlling the quadrotor are been paid attention to in the presence of external disturbances and uncertainties. First, unfalsified control method is introduced which is a data-driven and model-free approach in the field of adaptive control. Next, model of the quadrotor and unfalsified control design for the quadrotor are presented. Second, design of a control bank consisting of four proportional integral derivative controllers and a sliding mode controller is carried out.
Findings
A particular innovation on an unfalsified control algorithm in this paper is use of a generalized cost function in the hysteresis switching algorithm to find the best controller.
Originality/value
Finally, the performance and robustness of the designed controllers are investigated by simulation studies in various operating conditions including reference trajectory changes, facing to wind disturbance, uncertainty of the system and changes in payload, which show acceptable performances.
Neural adaptive control of two-link manipulator with sliding mode compensation
