A DSC method for strict-feedback nonlinear systems with possibly unbounded control gain functions

A DSC method for strict-feedback nonlinear systems with possibly unbounded control gain functions

Lv, Maolong (TU Delft Team Bart De Schutter; Air Force Engineering University)
Wang, Ying (Air Force Engineering University)
Baldi, S. (TU Delft Team Bart De Schutter)
Liu, Zongcheng (Air Force Engineering University)
Wang, Zutong (National Defence University)

2018

In dynamic surface control (DSC) methods, the control gain functions of systems are always assumed to be bounded, which is a restrictive assumption. This work proposes a novel DSC approach for an extended class of strict-feedback nonlinear systems whose control gain functions are continuous and possibly unbounded. Appropriate compact sets are constructed in such a way that the trajectories of the closed-loop system do not leave these sets, therefore, in these sets, maximums and minimums values of the continuous control gain functions are well defined even if the control gain functions are possibly unbounded. By using Lyapunov theory and invariant set theory, semi-globally uniformly ultimately boundedness is analytically proved: all the signals of closed-loop system will always stay in these compact sets, while the tracking error is shown to converge to a residual set that can be made as small as desired by adjusting design parameters appropriately. Finally, the effectiveness of the designed method is demonstrated via two examples.

Adaptive neural control
Dynamic surface control
Invariant set theory
Robust control

http://resolver.tudelft.nl/uuid:0795e7df-2fdc-4015-a34a-aee394e86cc0

https://doi.org/10.1016/j.neucom.2017.09.082

2019-12-19

0925-2312

Neurocomputing, 275, 1383-1392

Accepted Author Manuscript

Institutional Repository

journal article

© 2018 Maolong Lv, Ying Wang, S. Baldi, Zongcheng Liu, Zutong Wang