Cooperative Tracking Control of Nonlinear Multiagent Systems Using Self-Structuring Neural Networks

Abstract

This paper considers a cooperative tracking problem for a group of nonlinear multiagent systems under a directed graph that characterizes the interaction between the leader and the followers. All the networked systems can have different dynamics and all the dynamics are unknown. A neural network (NN) with flexible structure is used to approximate the unknown dynamics at each node. Considering that the leader is a neighbor of only a subset of the followers and the followers have only local interactions, we introduce a cooperative dynamic observer at each node to overcome the deficiency of the traditional tracking control strategies. An observer-based cooperative controller design framework is proposed with the aid of graph tools, Lyapunov-based design method, self-structuring NN, and separation principle. It is proved that each agent can follow the active leader only if the communication graph contains a spanning tree. Simulation results on networked robots are provided to show the effectiveness of the proposed control algorithms.

Distributed adaptive fault‐tolerant leader‐following formation control of nonlinear uncertain second‐order multi‐agent systems

“…wherẽi ≜ i − −̄i andṽ i ≜ v i −v ,̄i are the constant desired relative positions between the agents i and j; z i j are the constant design gains to be defined later in (31); , γ, , and are the positive constants defined in Lemma 3; sgn(·) is the sign function defined to take zero value at zero;…”

“…where i is defined in Lemma 1,k i is defined in the proof of Lemma 1, and k ij and k ji are positive constants denoting the weights on the intercommunication graph . Note that the distributed gains z ij given in (31) are the ith row and jth column entries ofΨ defined in Lemma 2. Therefore, ∑ ∈N i z i (̃i + γṽ i ) =Ψ̃+ γΨṽ, wherẽis the column stack vector of̃i ≜ i −̄i,̄i is the unknown constant desired relative position between the leader and agent i, andṽ is the column stack vector ofṽ i ≜ v i − v 0 .…”

Distributed secondary control for droop-controlled autonomous microgrid

“…The distributed control structure is suitable for the sparse communication network and easy to achieve the plug-and-play function. Motivated by the flexibility and computation efficiency of networked multi-agent systems (MAG), the MAG-based distributed cooperative control has received much attention in recent years [6][7][8][9]. This kind of control strategy can be applied in the secondary control level of microgrids.…”

Decentralized control for stabilization of nonlinear multi-agent systems using neural inverse optimal control

“…However most of them uses continuous time feedforward neural networks [4], [5], [7], [8] and usually require a lot of information about the topology. Instead, the scheme proposed in this paper is based on discrete-time recurrent neural networks and does not need to know the mentioned topology.…”

“…However, artificial neural networks, through their multiple application, have been demonstrated to be ideal for modeling complex nonlinear systems, due to their approximation capabilities, easy implementation, robustness against noise and online training, which make them very adequate for modeling complex nonlinear systems [19]. Today, there is a growing interest in the application of artificial neural networks for the design of multi-agent controllers [7], [20], [8] and references therein. However, most of such controllers have been designed for continuous-time systems, the discrete case has not been treated with the same depth, and however for real-time implementations it is very important to consider this case.…”