Practical Finite-Time Command-Filtered Adaptive Backstepping With Its Applications to Quadrotor Hovers

Area of Science:

• Control Systems Engineering
• Nonlinear Dynamics
• Robotics

Background:

• Uncertain nonlinear systems present significant control challenges.
• Existing adaptive backstepping methods often rely on complex components like neural networks or fuzzy logic systems.
• Addressing nonparametric uncertainties and external disturbances is crucial for robust control.

Purpose of the Study:

• To introduce a novel practical finite-time command-filtered adaptive backstepping (PFTCFAB) control method.
• To develop a PFTCFAB strategy that avoids the need for neural networks or fuzzy logic systems.
• To ensure finite-time stability for tracking and estimation errors in uncertain nonlinear systems.

Main Methods:

• Design of novel function adaptive laws for direct estimation of unknown nonlinearities and disturbances.
• Development of practical finite-time command filters to generate adaptive laws.
• Integration of PFTCFAB controllers and command filters with finite-time Lyapunov stability analysis.

Main Results:

• The proposed method effectively handles nonparametric unknown nonlinearities and external disturbances without NNs or FLSs.
• Finite-time stability of system tracking errors and filter estimation errors is guaranteed.
• Experimental validation on a quadrotor hover system demonstrates the strategy's effectiveness.

Conclusions:

• The presented PFTCFAB control method offers a simplified and more reliable approach to controlling uncertain nonlinear systems.
• The technique successfully addresses system uncertainties and disturbances, achieving finite-time stability.
• The study highlights the potential of command filter techniques in enhancing adaptive control strategies.