Search

This article investigates distributed optimization of high‐order nonlinear multi‐agent systems (MASs) with disturbance under switching topologies. First, we propose a class of new distributed control algorithms for high‐order MASs under switching topologies, such that all agents asymptotically reach consensus at the unique minimizer of the sum of cost functions. Then, by constructing disturbance observer, the proposed algorithms are extended to address distributed optimization problems for high‐order nonlinear MASs with disturbance under switching topologies. In particular, both of new presented methods do not require optimal reference signal generator, which can save the computational resources of systems. In final, simulations are conducted to validate the theoretical results. [ABSTRACT FROM AUTHOR]

Existing distributed finite‐time optimization algorithms for continuous‐time multi‐agent systems require either undirected graphs or weight balanced digraphs, and distributed finite‐time optimization problems for weight unbalanced digraphs are still great challenging. Thus, the distributed finite‐time optimization problems for continuous‐time multi‐agent systems with strongly convex local cost functions are investigated under directed graphs in this article. First, a distributed finite‐time gradient estimator is constructed by using non‐smooth analysis and algebraic graph theory, then distributed finite‐time optimization algorithms and piecewise distributed finite‐time optimization algorithms are proposed based on the designed gradient estimator. The new proposed distributed finite‐time optimization algorithms which only require strongly connected graphs relax the balanced requirement. Furthermore, the communication bandwidth of systems could be saved by deploying the proposed piecewise distributed finite‐time optimization algorithms since the information exchange in the optimization process is reduced. Finally, simulation examples are given to verify the effectiveness of proposed distributed optimization algorithms. [ABSTRACT FROM AUTHOR]

In this paper, we are concerned with distributed finite-time optimisation of multi-agent systems with unknown high-frequency gain signs under switching digraphs. First, a distributed finite-time gradient estimator-based optimal signal generator is constructed by adopting finite-time stability theory and graph theory. Then, considering the existence of unknown high-frequency gain signs in multi-agent systems, we propose new distributed asymptotic optimisation algorithms via the Nussbaum-type function. Additionally, in order to accelerate convergence, the novel distributed finite-time optimisation algorithms are presented. Both of the newly proposed algorithms can realise optimal consensus of multi-agent systems with unknown high-frequency gain signs under switching digraphs. Finally, two simulation examples are illustrated to verify the proposed algorithms. [ABSTRACT FROM AUTHOR]

In this article, we explore the possibility of designing a continuous control algorithm for time‐varying networks of high‐order nonlinear agents with input saturation. A class of neural network‐based (NN) distributed control algorithms are proposed for the consensus of nonlinear agents with input saturation and nonidentical unknown control directions (UCDs) under uniformly quasi‐strongly δ‐connected graphs. An NN is used to cancel the unknown part of agent dynamics due to the well known NN function approximation property. Also, an adaptive smooth term is employed to compensate for the bounded disturbance, the approximation error. Moreover, an auxiliary system is designed to counteract the effect of input saturation. Furthermore, the control law is augmented with the Nussbaum‐type function to obtain an estimation of the UCDs. It is proven rigorously that the consensus of nonlinear agents with input saturation and nonidentical UCDs can be guaranteed, and the boundedness of all signals of closed‐loop systems is ensured. Finally, simulation examples are given to show the effectiveness of our theoretical results. [ABSTRACT FROM AUTHOR]

Existing results for cooperative control of high-order agents mainly employ the Nussbaum-type function to cope with unknown control directions and mostly require an assumption that the control directions are identical and unknown. This paper proposes a class of algorithms with nonlinear PI functions to relax such an assumption and make them suitable for nonidentical unknown control directions. It is proven that if the distributed nonlinear PI functions are suitably selected, the proposed algorithms can achieve consensus for high-order agents under strongly connected topologies and switching topologies with a jointly strongly connected basis (JSCB). Furthermore, we extend the consensus results to the case of time-varying topologies described by $\delta $ -connected, continuous graphs. As a special case, the consensus of high-order agents under the directed graph having a spanning tree is also investigated. Finally, illustrative simulations are presented to indicate the efficiency of the proposed algorithms. [ABSTRACT FROM AUTHOR]

This paper solves the adaptive consensus problem for first-order linearly parameterized agents with completely nonidentical unknown high-frequency gain signs under directed graphs. A new class of Nussbaum-type function-based algorithms are proposed to handle the unknown high-frequency gain signs adaptively and cooperatively. It is shown that if the underlying topology is a fixed graph with strongly connected or switching topologies having a jointly strongly connected basis, the first-order linearly parameterized agents with nonidentical unknown high-frequency gain signs can achieve asymptotic consensus. Finally, the effectiveness of proposed algorithms are verified by one simulation example. [ABSTRACT FROM AUTHOR]

This paper addresses the coordinated tracking problem for linear multiagent systems with input saturation and stochastic disturbances. The objective is to construct a class of tracking control laws that achieve consensus tracking in the absence of disturbances, while guaranteeing a bounded variance of the state difference between the follower agent and the leader in the present of disturbances, under the assumptions that each agent is asymptotically null controllable with bounded controls (ANCBC) and the network is connected. By using the low gain feedback technique, a class of tracking control algorithms are proposed, and the coordinated tracking problem is solved through some routine manipulations. Finally, numerical examples are provided to demonstrate the theoretical results. [ABSTRACT FROM AUTHOR]

The problem of robust global consensus tracking of linear multiagent systems with input saturation and input-additive uncertainties is investigated in this paper. By using the parametric Lyapunov equation approach and an existing dynamic gain scheduling technique, a new distributed nonlinear-gain scheduling consensus-trackining algorithm is developed to solve this problem. Under the assumption that each agent is asymptotically null controllable with bounded control, it is shown that the robust global consensus tracking can be achieved under the undirected graph provided that its generated graph contains a directed spanning tree. Compared with the existing algebraic Riccati equation approach, which requires the online solution of a parameterized algebraic Riccati equation, all the parameters in the proposed nonlinear algorithm are offline determined a priori. Finally, numerical examples are provided to validate the theoretical results. Copyright © 2016 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

This paper solves the robust output synchronization problem for a group of linear agents subject to constant disturbances on directed graphs. A new distributed control protocol with integral action is first proposed with the assumption that each agent has access to its own state. Then, an observer-based implementation of the control protocol is illustrated for the case that the agents do not have access to their states. Compared with existing methods, the new distributed protocol does not require exchanging of the controller states or output states among neighboring agents. Finally, simulation results of a group of low-speed experimental unmanned aerial vehicles show the effectiveness of the proposed methods. Copyright © 2017 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

This paper proposes a new class of nonlinear protocols for fixed-time consensus in networked agents with directed and intermittent communications. First of all, based on the assumption of a common positive dwell time for all successful links, it is theoretically shown that the asymptotical consensus can be achieved if the interaction topology with some fixed length of time has a spanning tree. Then, if the length sum of time intervals, over which the directed interaction topology contains a spanning tree, is larger that a threshold value, which is independent of initial conditions, the system will solve a fixed-time consensus problem. Furthermore, in light of the mirror operation principle of graph theory, the proposed protocol is extended to solve the fixed-time consensus problem if a common positive dwell time for all active links is strongly connected without the detailed balanced condition. Finally, the effectiveness of the theoretical results is illustrated by simulation results. [ABSTRACT FROM AUTHOR]

This paper addresses the exponential consensus problem of single-integrator agents with saturated protocols on directed graphs. By employing an integral Lyapunov function, the exponential consensus problem of single-integrator agents is solved under the directed graph with strongly connected or a spanning tree. The main contribution is that under the directed graph, some conditions for exponential consensus with saturated protocols are first obtained. Finally, two examples are used to illustrate the effectiveness of the theoretical results. [ABSTRACT FROM AUTHOR]

This paper investigates the semiglobal synchronization problem for a group of agents with input saturation under directed interaction topology, where each agent is modeled as a generic linear system rather than the single-integrator or double-integrator dynamics. The main result is the construction of a feedback coupling gain that achieves semiglobal synchronization if all the agents have identically saturated linear dynamics, which can be of any order. It is shown that the coupling gain obtained via parametric Lyapunov equations can semiglobally synchronize any directed network provided that the interaction topology has a directed spanning tree. Furthermore, due to the use of parametric Lyapunov equations, a convergence rate is analytically obtained. Finally, a simulation example is provided to demonstrate the effectiveness and advantages of our theoretical findings. [ABSTRACT FROM AUTHOR]

Abstract: This paper investigates the synchronization problem for a group of agents with identical linear dynamics subject to input saturation. Two classes of linear systems, asymptotically null controllable with bounded control (ANCBC) systems and double-integrator systems, are studied. For ANCBC systems, it is shown that a linear protocol with the control gain obtained via parametric Lyapunov equations can semiglobally synchronize the undirected topology provided that its augmented directed topology has a spanning tree. For a special case of ANCBC, the double-integrator systems, it is established that if the augmented directed topology has a spanning tree, then in the presence of input saturation, using linear protocols with positive control gains can achieve global synchronization. Two numerical examples are given to demonstrate the effectiveness of the theoretical results. [Copyright &y& Elsevier]