Fuzzy Observer-Based Finite-Time Adaptive Formation Control for Multiple QUAVs With Malicious Attacks

This article focuses on the finite-time formation control problem for multiple quadrotor unmanned aerial vehicles (QUAVs) with malicious attacks, and presents a finite-time fuzzy adaptive output-feedback control scheme. First, the positional and angular velocities are estimated by developing the fuzzy state observer to replace actual values for controller design. Second, the problem of "computational complexity" is avoided and the effect of filtered error is eliminated by introducing the finite-time command filtered technique and constructing the error compensation mechanism, respectively. Meanwhile, the adaptive parameters are used to estimate the boundaries of malicious attack signals, overcoming the challenge of requiring bounds for attack signals in the backstepping design process. Based on the finite-time stability theory, it is proven that all signals are bounded in the multiple QUAVs system, and the formation tracking errors can converge to a sufficiently small neighborhood near the origin in a finite time. Finally, the validity of the algorithm is verified by a simulation example.