Prescribed-time coordinated control for fixed-wing UAV close formation

During close formation flight of fixed-wing UAVs, the wingman aircraft is subjected to interference from the wake vortex of the leader aircraft, increasing the difficulty of designing the wingman's controller. To address this issue, this paper first proposes a modeling method for aerodynamic coupling in close formation, establishes a mathematical model of the wake vortex for highly swept-back wing UAVs, and studies the mechanism of aerodynamic coupling in close formation. Subsequently, the concept of prescribed-time control is introduced into Incremental Nonlinear Dynamic Inversion (INDI) control, enabling the controller to achieve rapid convergence while maintaining strong robustness. Based on this control method, the inner-loop controller for the wingman aircraft is designed to ensure accurate tracking of command signals under aerodynamic disturbances. Since the incremental nonlinear dynamic inversion controller requires angular rate and angular acceleration signals for controlling the inner loop, but the angular rate signals obtained from sensors in practice contain measurement noise, traditional differentiation methods amplify noise and fail to provide accurate angular acceleration signals. To resolve this, the concept of predefined-time control is incorporated into the tracking differentiator, designing a modified tracking differentiator that achieves simultaneous signal tracking and differentiation extraction under noisy conditions, with both robustness and rapid response. The stability of the proposed controller is proven using the Lyapunov theorem, and digital simulations of the entire closed-loop system are conducted. Simulation results demonstrate that the designed controller achieves the expected control effect and meets the attitude control requirements of the wingman aircraft during fixed-wing UAV close formation flight. Comparison with control methods from other literature shows that in close formation scenarios, the controller designed in this paper exhibits faster convergence, smaller steady-state error, and stronger robustness.