Neural adaptive sliding mode control for insulator cleaning robots with actuator faults and prescribed performance

This article investigates the trajectory tracking control of an insulator cleaning robot confronted with model inaccuracy, external disturbances, actuator faults, and prescribed performance constraints. By leveraging the universal approximation capability combined with its inherent robustness, a neural networks-based adaptive sliding mode controller is designed to handle the model inaccuracy and disturbances. In a bid to ensure the operational safety of the catenary and cantilever system, a fault-tolerant mechanism is integrated with a prescribed performance function, thereby addressing both actuator faults and control performance requirements. The proposed control scheme guarantees that the trajectory tracking error is Uniformly Ultimately Bounded (UUB) and the performance can be pre-specified even in the presence of actuator faults and disturbances. The stability of the proposed control system is rigorously proven via the Lyapunov method, and the effectiveness of the proposed approach is demonstrated through numerical simulations purely.