A Data-Driven Distributed Adaptive Control Approach for Nonlinear Multi-Agent Systems

Abstract

In this paper the distributed leader-follower consensus tracking problem is investigated for unknown nonlinear non-affine discrete-time multi-agent systems. Via a dynamic linearization method both for the agent system and the local ideal distributed controller, a distributed adaptive control scheme is proposed in this paper using the Newton-type optimization method. The proposed approach is data-driven since only the local measurement information among neighboring agents is utilized in the control system design. The consensus tracking stabilities of the proposed approach are rigorously guaranteed in the cases of fixed and switching communication topologies. The simulations are conducted to verify the effectiveness of the proposed approach. INDEX TERMS Dynamic linearization, data-driven control, adaptive control, multi-agent systems, consensus tracking.

Prescribed-Time Average Consensus Through Data-Driven Leader Motion

“…As a first step, we study under which conditions on a i and b i , the multi agent system under the control input (25) with algorithmic structure (29) reaches an agreement. We follow the detailed analysis adopted in [40].…”

“…As a final goal of this paper, we assume that the leader has poor a-priori knowledge of the network, and it is forced to build the algorithm structure through a data-based learning procedure, which is run in an initial stage of the system evolution. Indeed, in the last years an ever-increasing research thrust has been put into the design of model-free and datadriven control strategy [24], [25]. [26].…”

HDP Algorithms for Trajectory Tracking and Formation Control of Multi-Agent Systems

“…(1) and step 12 uses Eq. (17). The control update of the agent i, represented by steps 16-19, is the third function block.…”

“…A promising solution to solve the online control problem with plant uncertainties is the adaptive dynamic programming (ADP) method [14], [15]. This approach employs a 'forward-in-time' mechanism that looks for an optimal control policy by successively adapting two parametric structures, i.e., an action network and a critic network, to approximate the solution of the Hamilton-Jacobi-Bellman (HJB) equation [16], [17].…”

Data-Driven Containment Control for a Class of Nonlinear Multi-Agent Systems: A Model Free Adaptive Control Approach

“…MFAC has also been employed in the coordination of MASs [34][35][36][37][38][39][40][41][42]. MFAC is first used to address the consensus problem for a group of unknown heterogenous nonlinear MASs in [34].…”