Safe collaborative control for UAH formation based on improved SODO and tracking differentiator under dynamic boundary constraints

This paper investigates the problem of safe collaborative control for unmanned aerial helicopter (UAH) formation under dynamic boundary constraints and external disturbances. Firstly, an improved boundary protection algorithm (IBPA) is proposed to tackle the issue on dynamic boundary constraints of the UAH formation, which instantaneously constructs the dynamic security trajectories for the follower UAHs. Secondly, for each UAH, an improved second-order disturbance observer (SODO) is proposed by incorporating state prediction error to achieve efficient and accurate estimation under lower gain. Thirdly, to achieve safe efficient tracking, a dual-layer control strategy is presented: the first layer introduces a virtual UAH and presents a force fast tracking differentiator (FFTD), effectively smoothing the severe deflection generated when the dynamic safe trajectories approach the constraint boundaries and suppressing high-frequency oscillation within the safe boundaries; the second layer adopts the trajectory switching algorithm (TSA) and proposes the controller design based on backstepping method, which further improves the tracking accuracy and reduces computational complexity. Fourthly, by borrowing Lyapunov theory, the stability for overall closed-loop system composed of estimation errors and tracking errors is analyzed, thus establishing the method of deriving observer gains and controller ones. Finally, the effectiveness and superiority of the proposed scheme is verified by resorting to simulated results.