Abstract
This study considers the problem of finite-time attitude control for quadrotor unmanned aerial vehicles (UAVs) subject to parametric uncertainties, external disturbances, input saturation, and actuator faults. Under the strong approximation of radial basis function neural networks (RBFNN), an adaptive finite-time NN observer is first presented to obtain the accurate information of unavailable angular velocity. More importantly, an adaptive mechanism to adjust the output gain of the fuzzy logic system (FLS) is developed to avoid the selection of larger control gains, and can even work well without the prior information on the bound of the lumped disturbance. Based on the nonsingular fast terminal sliding mode manifold, a novel switching control law is designed by incorporating the adaptive FLS and fast continuous controller in order to remove the undesired chattering phenomenon and solve the negative effects induced from the parametric uncertainty, external disturbance, and actuator fault. To deal with the input saturation, an auxiliary system is constructed. The rigorous theoretical analysis is given to prove that all the signals in the closed-loop system are uniformly bounded, and tracking errors converge into bounded neighborhoods near the origin in finite time. Moreover, the issue of selecting control parameters is analyzed in detail. Last but not least, the comparative simulation results show the validity and feasibility of the proposed control framework.
Introduction
Quadrotor Unmanned Aerial Vehicles (UAVs) are inherently unstable, highly nonlinear systems that are exceptionally sensitive to environmental factors. While numerous control strategies have been proposed, practical implementation often faces three critical challenges: heavy dependency on precise angular velocity sensors (gyroscopes), vulnerability to external disturbances and actuator faults, and the "chattering" phenomenon associated with robust control methods like Sliding Mode Control (SMC).
This research addresses these gaps by developing a "velocity-free" controller. Unlike traditional systems that rely on potentially noisy or failing gyroscopes, the proposed framework guarantees high-precision attitude tracking and finite-time convergence—ensuring the drone reaches its desired orientation within a predictable timeframe even under adverse conditions.
Methodology
The proposed control architecture integrates several intelligent components to handle the complexities of real-world flight:
- Adaptive Finite-Time NN Observer (AFTNNO): Utilizing a Radial Basis Function Neural Network (RBFNN), the system estimates angular velocity without direct sensor measurements. This adaptive observer learns to compensate for model uncertainties in real-time.
- Fuzzy Logic System (FLS): An FLS approximates "lumped disturbances," which include wind gusts, parametric uncertainties, and actuator faults. A key innovation is the adaptive gain mechanism, which removes the need for prior knowledge of disturbance bounds.
- Nonsingular Fast Terminal Sliding Mode (NFTSM): This core logic ensures rapid convergence of tracking errors while avoiding mathematical singularities.
- Anti-Saturation Auxiliary System: This component manages physical motor limits, preventing "wind-up" when required torque exceeds hardware capabilities.
The framework was validated through extensive MATLAB/Simulink simulations, subjecting the model to time-varying wind, payload shifts, and simulated motor failures.
Results and Discussion
The simulation results demonstrate that the proposed "intelligent" approach significantly outperforms standard benchmarks like Proportional-Derivative (PD) and Dynamic Surface Control (DSC).
- Precision: The proposed method achieved an Integral Squared Error (ISE) of approximately 0.0099, representing a nearly six-fold improvement over the PD controller (ISE 0.0576).
- Efficiency: Total Energy Consumption (TEC) remained low (~0.1196), proving that advanced robustness does not necessarily come at the cost of battery life.
- Robustness: The controller successfully "masked" hardware failures, maintaining a stable attitude where traditional methods exhibited significant oscillations.
A major advantage of this design is the elimination of "chattering." By combining NFTSM with a continuous fuzzy compensator, the control signals are smoothed, which reduces mechanical wear on motors and improves overall energy efficiency.
Conclusion
This study presents a robust, velocity-free control framework that effectively manages the nonlinearities and uncertainties of quadrotor flight. By moving away from sensor dependency and providing finite-time stability guarantees, the proposed method offers a practical path toward more resilient autonomous systems. Future work will focus on extending the logic to handle "stuck faults" and validating the framework on physical hardware to confirm its real-time performance.
Summary updated