LAUSR

In this article, we propose a redundant fault-tolerant and boundary constraint control in the framework of adaptive method, the neural network approximation, and the barrier Lyapunov function (BLF) with a relaxed initial condition. The actuator failures are compensated by a combination of adaptive methods and redundant actuators when some actuators suffer from partial or even total loss of effectiveness. The radial basis function’s neural network structure is introduced to strengthen the adaptivity in various orientations of the hose and other additionally unmodeled dynamics. To maintain the boundary deflection within a predefined open set after a constraint time, a novel asymmetrical and time-varying BLF is constructed by applying a shifting function to transform the original state into a new state with zero value initially. The performance of the developed adaptive control is demonstrated by numerical simulations. Note to Practitioners—This article is motivated by the limited performances in the existing control designs for flexible unmanned aerial refueling hose systems with failed actuators and boundary constraints. This article considers a redundant actuator case to solve time-varying and partially and totally failed actuator failures that are not settled by adaptive and Nussbaum-based single control. Unlike conventional barrier Lyapunov functions (BLFs), in this article, we resort to a shifting function and propose a novelly asymmetric and time-varying BLF, which is well defined initially and capable to address a deferred constraint control problem. The control design and stability analysis of the actuated system is a Lyapunov-based method rather than relying on semigroup theory or functional analysis, which makes the developed method more engineering orientated. The proposed control strategy is tested to illustrate performances in numerical simulations with the finite difference method.