Self-Triggered Distributed Formation Control of Fixed-Wing Unmanned Aerial Vehicles Subject to Velocity and Overload Constraints

In this study, the self-triggered distributed formation control problem is considered for a group of fixed-wing unmanned aerial vehicles (UAVs) under unknown wind disturbances subject to velocity and overload constraints. A more general kinematic model of UAVs involving the position, velocity, velocity yaw angle, velocity pitch angle, velocity roll angle, acceleration, overload, and wind disturbances is employed. Based on the sampled local relative position information, a novel distributed control scheme with an inner-outer loop structure and a self-triggered sampling mechanism are developed. Velocities and overloads of the controlled fixed-wing UAVs are maintained within desired ranges by introducing appropriate saturations to the loops. The nonlinear small-gain method and differential inclusion theories are used as fundamental tools to establish the stability of the closed-loop system. It is shown that the formation control objective can be achieved if a spanning tree condition is satisfied by the interconnection topology. Finally, the effectiveness and superiority of the proposed method are illustrated by simulation and experimental studies with subsonic UAVs. Note to Practitioners—This study is motivated by the emerging need for the self-triggered distributed formation control of fixed-wing UAVs. The inherent difficulty of distributed formation control for a group of fix-wing UAVs with unknown wind disturbances subject to various constraints (e.g., velocity constraints, overload constraints) is intensified by a large number of communication resources and coupling constraints. Based on the constructed kinematic model of UAVs, the presented self-triggered distributed control strategy with an inner-outer loop structure and a self-triggered sampling mechanism can achieve the formation control objective and reduce the use of communication resources, making a trade-off between formation performances and communication resources. The proposed strategy is universal and versatile, as it can be applied to different UAVs formation control frameworks and also implemented in heterogeneous multi-agent systems. Preliminary experiment tests suggest that this approach is feasible and the model does consider unknown wind disturbances. In future research, our modeling approach will integrate dynamic uncertainty to improve the proposed strategy.