Feedback Linearization With Intelligent Disturbance Observer For Autonomous Quadrotor With Time-varying Disturbance

The presence of disturbances during flight may destabilize the quadrotor control and could compromise the designated mission. This paper proposes an improved quadrotor flight control in the presence of wind disturbances where the performance will be more robust in many flight conditions. This is achieved by integrating artificial intelligence (AI) technique with disturbance observer-based feedback linearization to improve the disturbance approximation and compensation. The AI technique via radial basis function neural network (RBFNN) is implemented to compensate the bounded estimation error produced by the disturbance observer. The weights of the neural network are tuned online with no prior training required. Simulation results demonstrate the effectiveness and feasibility of the proposed technique.
Keywords: Disturbance observer, feedback linearization, neural network, quadcopter