A source seeking algorithm with application to a quadcopter model

Abstract

A novel extremum seeking (ES) algorithm is developed and its performance is tested on a quadcopter type unmanned aerial vehicle (UAV) model. The motivation of the paper is to develop an ES algorithm that results no oscillations in the system behaviour during the search process. Via the proposed algorithm, the quadcopter searches autonomously a signal source/target without knowing its location a priory. The controller does not use any position or GPS measurements. The presented control algorithm is composed of two phases: In the first phase, the search algorithm determines required translational acceleration values of the UAV. In the second phase, a PID-type controller is utilized to control roll and pitch angles of the UAV to maintain computed acceleration values via the search algorithm. Performance of the controller is evaluated via simulation studies.