Control of Quadrotor Based on RBF Neural Network Adaptive Fast Terminal Sliding-Mode Strategy

Abstract I The purpose of this paper is to propose a fast adaptive terminal sliding mode control method for quadrotor unmanned aerial vehicles (UAVs) under dynamic uncertainties and external interference. Firstly, complete mathematical dynamics model UAV presented. Secondly, robust dual-loop nonlinear law the velocity attitude tracking designed. Through introduction equivalent inputs, channels can be decoupled stabilized controllers designed separately. For altitude control, coupled are ensure based on stability control. loop adopts RBF neural network compensate disturbance uncertainty, adjust its gain in real time, suppress chattering problems. Finally, experimental results show that controller not only interference, but also achieve accurate desired flight command.