Work

Finite-Time Fault-Tolerant Attitude Control for Quadrotor via Adaptive Sliding Mode Control and RBFNN Observer

Year: 2026
Type: article
Abstract: This paper presents an observer-assisted barrier adaptive sliding mode controller for quadrotor attitude tracking under nonlinear uncertainties, external disturbances, and matched actuator faults. An RBFNN observer estimates the nominal nonlinear dynamics and disturbances using linear and fractional-power error feedback, whereas the barrier term compensates for the remaining estimation residual and fault effect. An initial-error compensation term initializes the sliding variable at zero. Under bounded-fault and neural-approximation assumptions, Lyapunov analysis establishes bounded neural weights, finite-time entry of the observer error into an explicit neighborhood, forward invariance of the prescribed sliding domain, and bounds on the barrier gain and attitude-tracking error. Comparative simulations against a published fault-tolerant controller, together with combined-stress tests and mechanism-specific ablations, show improved post-fault tracking with the simulated control torques remaining within the prescribed limits and verify the respective roles of the modified sliding surface and observer compensation.
Source: Actuators
Language: English
Open Access status: gold
Estimated APC paid: $2,896
OpenAlex ID: W7235206631
FWCI:
Cites: 36
Cited by:
Related to: 10