Learning-based safety-guaranteed sliding mode affine formation maneuver control of quadrotors vulnerable to...

This article studied the actor-critic learning scheme for safe leader-follower affine formation maneuver control of networked quadrotors under external disturbances, sensor deception attacks, and injection attacks on the actuators. The followers aim to track formation maneuvers such as scaling, shearing, translation, and rotation determined by the leaders. Motivated by increasing safety and performance requirements during formation maneuvering, the dynamic states of the quadrotors are constrained within prescribed safety constraints. A barrier Lyapunov function is employed to ensure that the safety constraints are not violated. Then, a distributed sliding mode control with actor-critic learning is formulated to facilitate accurate leader-follower affine formation maneuvers and reject malicious cyber-attack signals. The input gains that appear due to the attacks might corrupt the control direction. The Nussbaum gain function is coupled to the controller to tackle this problem. The actor system estimates the uncertain dynamics and malicious attack signals, while the critic network evaluates the control performance through the estimated long-term performance index. The overall stability of the closed-loop system has been proven to be bounded using the Lyapunov stability theorem. Finally, simulation results demonstrate the capability of the presented control method.