Premium
Neuro‐adaptive distributed output‐feedback containment control for multiagent systems with nonstrict‐feedback nonlinear dynamics and input constraints
Author(s) -
Peydayesh Amirhossein,
Arefi Mohammad Mehdi
Publication year - 2021
Publication title -
international journal of robust and nonlinear control
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.361
H-Index - 106
eISSN - 1099-1239
pISSN - 1049-8923
DOI - 10.1002/rnc.5416
Subject(s) - backstepping , control theory (sociology) , nonlinear system , controller (irrigation) , computer science , a priori and a posteriori , observer (physics) , bounded function , containment (computer programming) , adaptive control , multi agent system , state observer , strict feedback form , control engineering , control (management) , mathematics , engineering , artificial intelligence , mathematical analysis , philosophy , physics , epistemology , quantum mechanics , agronomy , biology , programming language
In this article, distributed adaptive controllers are designed for containment control of multiagent systems (MASs) with high‐order nonlinear dynamics in nonstrict‐feedback form. To broaden the application horizon, the agents' dynamics are considered heterogeneous, and the control saturation is taken into account. Moreover, the agents' dynamics, as well as their full state information, are assumed unknown. An adaptive observer is designed in the first step for each agent to estimate the agents' states using available output measurements. An adaptive backstepping controller is then integrated with the observer to solve the containment control problem. The computational complexity of the controller is reduced by adopting the command filtered adaptive backstepping approach to the containment control of MASs. The effect of the input saturation is also compensated by employing an auxiliary system throughout the design procedure. The proposed control approach guarantees that the states of all agents are cooperatively semiglobally uniformly ultimately bounded and containment errors converge to a small neighborhood of the origin. Finally, the effectiveness of the presented control algorithm is confirmed by a simulation example.