关键词: 四旋翼无人机, 编队控制, 安全控制, 预设性能, 神经网络, 干扰观测器

Abstract: Quadrotor UAVs formation suffers from the problems of model uncertainties, unknown external disturbances and collision between UAVs. In this paper, a safety control scheme is proposed based on prescribed performance. Firstly, the inequality constraint problem which will prevent collision between UAVs is transformed into the unconstrained problem according to the Prescribed Performance Function (PPF) and the error transfer function. To tackle model uncertainty, the neural network is used for approximation. The unknown approximation errors and the unknown external disturbances are treated as a compound disturbance, which is then estimated by the nonlinear disturbance observer. By using the transformed tracking errors and the values obtained by the observer, controllers are designed for the position and attitude subsystems, thus the collision between quadrotor UAVs is avoided. Then, the convergence of all the closed-loop system signals under the designed controller is proved by the Lyapunov method. Finally, numerical simulations verify the effectiveness of the proposed scheme.

Key words: quadrotor UAV, formation control, safety control, prescribed performance, neural network, disturbance observer