Adaptive fast sliding neural control for robot manipulator

Authors

BARIŞ ÖZYER

DOI

10.3906/elk-2001-129

Abstract

Robotic manipulators are open to external disturbances and actuation failures during performing a task such as trajectory tracking. In this paper, we present a modifed controller consisting of a global fast sliding surface combined with an adaptive neural network which is called adaptive fast sliding neural control (AFSNC) for a robotic manipulator to precise stable trajectory tracking performance under the external disturbances. The adaptive term is employedtoreduce uncertainties due to unmodeled dynamics. Trackingerror asymptoticallyconvergesto zero according to the Lyapunov stability theorem. Numerical examples have been carried on a planar two-links manipulator to verify the control approach efficiency. The experimental results show that the proposed control approach performs satisfactory trajectory tracking and tracks the desired trajectory in less time with reduced chattering effect compared to the other methods.

Keywords

Sliding mode control, adaptive control, neural network control, asmyptotic stability, robot manipulator

First Page

3154

Last Page

3167

Recommended Citation

ÖZYER, BARIŞ (2020) "Adaptive fast sliding neural control for robot manipulator," Turkish Journal of Electrical Engineering and Computer Sciences: Vol. 28: No. 6, Article 5. https://doi.org/10.3906/elk-2001-129
Available at: https://journals.tubitak.gov.tr/elektrik/vol28/iss6/5