Leader–follower synchronisation for networked Lagrangian systems with uncertainties: a learning approach

Abstract

This article addresses a leader–follower synchronisation problem of networked Lagrangian systems with uncertainties by an iterative learning control approach. The inherent properties of the systems are fully utilised in the controller design, and a directed acyclic graph is sufficient for communication among subsystems. The developed controller contains a proportional-plus-derivative (PD) term and two learning terms. The PD term drives the tracking error to zero, one learning term compensates for the model uncertainties, and the other one is used for disturbance rejection. It is shown that the synchronisation task can be achieved by the proposed controller, and all internal signals are either bounded or norm bounded. The theoretical results are supported by a numerical study.