Adaptive bipartite output consensus of nonlinear fractional-order multi-agent systems

Abstract

A distributed control approach is proposed for a class of fractional-order multi-agent systems with unknown nonlinearities under signed bipartite digraphs. Unlike the existing results, the follower's dynamics are studied with strict-feedback fractional-order form as well as a general class of actuator faults with unknown magnitude, pattern and occurrence time is studied. To provide a simple and efficient control strategy, first a novel command fractional-order filter based on the backstepping design is proposed such that the difficulty of calculation of the fractional derivative order of the virtual control laws is removed, as well as improving the bipartite output consensus accuracy. Then, to cope with dynamic's uncertainties, the proposed method is integrated with adaptive neural approximator and minimal learning parameter scheme which reduces communication loads. Besides, a distributed fault compensation protocol based upon the proposed command fractional-order filter and relative output information of neighbours' agents is extended to ensure bipartite output consensus, without relying on any global information of the singed digraph as well as any explicit fault detection mechanism. Finally, it is guaranteed that all the error signals within the closed-loop network system are converged into adjustable compact sets around the origin. The simulation results verify the validity of the presented control approach.

Keywords:

• Bipartite output consensus
• fractional-order systems
• strict feedback
• command fractional-order filter
• actuator faults
• distributed control

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data sharing is not applicable to this article as no new data were created or analysed in this study.

Additional information

Notes on contributors

Hadi Mahmoodi

Hadi Mahmoodi was born in Ahwaz, Iran, on 23 April 1995. He successfully received his B.S. degree in control engineering from Islamic Azad University, Najafabad Branch, Iran in 2016 and received the M.S. degree with distinction in control engineering from Islamic Azad University, Najafabad Branch, Iran in 2018. His research interests include adaptive and nonlinear control, fractional-order systems, reinforcement learning, multi-agent systems, and event-based control.

Khoshnam Shojaei

Khoshnam Shojaei was born in Esfahan, Iran, on 8 March 1981. He received his B.S. degree, M.S. degree, Ph.D. degree with distinction in Electrical Engineering from Iran University of Science and Technology (IUST) in 2004, 2007 and 2011, respectively. Currently, he is an associate professor in Najafabad Branch, Islamic Azad University. His research areas are adaptive control of nonlinear systems, control of autonomous robots including land, air and ocean vehicles, and navigation of mobile robots.