• [email protected]
  • +971 507 888 742
Submit Manuscript
SciAlert
  • Home
  • Journals
  • Information
    • For Authors
    • For Referees
    • For Librarian
    • For Societies
  • Contact
  1. Information Technology Journal
  2. Vol 12 (4), 2013
  3. 696-703
  • Issues
    Online First Current Issue All Issues
  • Information About
    Aims and Scope Editorial Board Guide to Authors Article Processing Charges
    Submit a Manuscript

Information Technology Journal

Year: 2013 | Volume: 12 | Issue: 4 | Page No.: 696-703
DOI: 10.3923/itj.2013.696.703
crossmark

Facebook Twitter Reddit Linkedin E-mail
Research Article

Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data

Yaping Liu
Wuxi Research Center of Environmental Science and Engineering, Wuxi City College, Wuxi 214122, China

Zhihai Wu
School of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China

ABSTRACT


The stochastic bounded consensus tracking problems of leader-follower multi-agent systems were investigated, where the control input of each agent can only use the information measured at the sampling instants from its neighbors or the virtual leader with a time-varying reference state and the measurements are corrupted by random noises. The probability limit theory and algebra graph theory were employed to derive the sufficient conditions guaranteeing the mean square bounded consensus tracking. Simulations were provided to demonstrate the effectiveness of the theoretical results.
PDF Abstract XML References Citation

Keywords


  • Consensus
  • multi-agent systems
  • time-varying reference state
  • sampled-data
  • measurement noises
  • coordination tracking

Article History

Received: October 21, 2012;   Accepted: December 26, 2012;   Published: January 22, 2013

How to cite this article

Yaping Liu and Zhihai Wu, 2013. Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data. Information Technology Journal, 12: 696-703.

DOI: 10.3923/itj.2013.696.703

URL: https://scialert.net/abstract/?doi=itj.2013.696.703

INTRODUCTION


In recent years, distributed coordinated control of multi-agent systems has attracted a great deal of attention. This is due to its broad applications in various fields including physics, biology, robotics and control engineering. The consensus problem plays an important role in the distributed coordinated control of multi-agent systems. Consensus means that each agent updates its own state based on the states of its local neighbors such that the states of all the agents converge to a common value. For most of the existing consensus protocols, the final common value to be achieved is a function of initial states of all the agents and is inherently a prior unknown constant. This is the so-called χ-consensus (Cortes, 2006), one of important forms of which is the average consensus (Saber and Murray, 2004). Saber and Murray (2004) considered the average-consensus problems of networks of first-order integrator agents with fixed and switching topologies. They proved that if the network is an instantaneous balanced and strongly connected digraph, then the average consensus can be achieved. Saber et al. (2007) provided a theoretical framework for the analysis of consensus protocols for multi-agent systems, an overview of basic concepts of information consensus in networks and the methods of convergence and performance analysis for the protocols. Li and Zhang (2009a) designed the optimal weights for guaranteeing the fast average consensus of networked multi-agent systems. More results can be seen in the survey papers (Murray, 2007; Ren et al., 2007) and the references therein.

In real applications, it is often required that all the agents converge to a desired common value evolving over time. This is the so-called consensus tracking problem, where all the agents exchange their own state information with their neighbors with the common goal of tracking a time-varying reference state available to only a portion of agents. Thus, the χ-consensus protocols are ineffective to directly deal with the consensus tracking of a desired time-varying reference state. To solve this problem, consensus protocols of leader-follower multi-agent systems have been proposed. Up to now, significant attention has been focused on the consensus tracking of leader-follower multi-agent systems (Jadbabaie et al., 2003; Ren and Beard, 2005). Jadbabaie et al. (2003) considered the nearest neighborhood principle. They proved that if all the agents are jointly connected with their leader, then their states converge to the state of the leader as time goes on. Ren and Beard (2005) extended the results of Jadbabaie et al. (2003) to the directed topology case and gave some more relaxed conditions.

Most researches on the consensus tracking of leader-follower multi-agent systems in the above literature assumed that each agent measures the states of its neighbors and the leader accurately. Obviously, this assumption is only an ideal approximation for real communication channels, since real networks are often in uncertain communication environments with various source noises, channel noises and sink noises. Thus, it is necessary to study the effects of measurement noises on the consensus tracking of leader-follower multi-agent systems (Kingston et al., 2005; Ren et al., 2005; Huang and Manton, 2007a, b, 2009, 2010; Xiao et al., 2007; Ma et al., 2008; Li and Zhang, 2009a, 2010; Hu and Feng, 2010). Recently, consensus of leaderless multi-agent systems with measurement noises has been widely studied (Kingston et al., 2005; Ren et al., 2005; Huang and Manton, 2007a, b, 2010; Xiao et al., 2007; Li and Zhang, 2009a, 2010). Kingston et al. (2005) and Ren et al. (2005) introduced the time-varying consensus gains and designed the consensus protocols based on a Kalman filter structure. They proved that when there is no noise, the protocols designed can ensure consensus to be achieved asymptotically. Xiao et al. (2007) considered the first-order discrete-time average consensus of multi-agent systems with additive input noises. They designed the optimal weighted adjacency matrix to minimize the static mean square error. Since the consensus gain and the adjacency matrix are time-invariant, the average state diverges with probability one as time goes on, even if the noises are bounded. The stochastic approximation type protocols with the decreasing consensus gains were employed to guarantee the mean square or almost sure consensus of the first-order discrete-time multi-agent systems with measurement noises (Huang and Manton, 2007a, b, 2010; Li and Zhang, 2010). Li and Zhang (2009a) investigated the average consensus of the first-order continuous-time multi-agent systems with measurement noises and obtained the convergence conditions guaranteeing the mean square consensus. However, compared with consensus of leaderless multi-agent systems with measurement noises, the consensus tracking of leader-follower multi-agent systems with measurement noises has attracted only a little attention (Ma et al., 2008; Huang and Manton, 2009; Hu and Feng, 2010). Huang and Manton (2009) considered the consensus of discrete-time leader-follower multi-agent systems with measurement noises. By using the stochastic Lyapunov analysis, they derived the sufficient conditions guaranteeing the stochastic approximation type protocols with the decreasing consensus gains to reach the mean square consensus. Ma et al. (2008) investigated the consensus of continuous-time leader-follower multi-agent systems with measurement noises. By employing the stochastic analysis and algebra graph theory, they obtained the sufficient conditions for the mean square consensus. Hu and Feng (2010) developed a distributed tracking control scheme with distributed estimators based on a novel velocity decomposition technology for a continuous-time leader-follower multi-agent system with measurement noises. They proved that the closed loop tracking control system is stochastically stable in mean square and the estimation errors converge to zero in mean square as well.

For most of the work in the above literature, continuous-time protocols were used for continuous-time agent dynamics and discrete-time protocols were used for discrete-time agent dynamics. However, due to the application of digital sensors and controllers, in many cases, though the system itself is a continuous process, only sampled-data at discrete sampling instants is available for the synthesis of control laws. Compared with full-state information transmission, which might result in high communication traffic, the transmission of the state information at the sampling instants can effectively save the bandwidth of networks and communication cost. Considering the broad applications of digital communication and control in distributed systems, it is of great significance to study the consensus of multi-agent systems based on sampled-data control (Hayakawat et al., 2006; Hu and Hong, 2007; Mohanarajah and Hayakawa, 2008; Ren and Cao, 2008; Cao and Ren, 2009; Gao et al., 2009; Li and Zhang, 2009b; Cao and Ren, 2010a, b; Gao and Wang, 2010a-c). Up to now, considerable efforts have been focused on the sampled-data consensus of leaderless multi-agent systems without measurement noises (Hayakawat et al., 2006; Mohanarajah and Hayakawa, 2008; Ren and Cao, 2008; Cao and Ren, 2009; Gao et al., 2009; Cao and Ren, 2010a, b; Gao and Wang, 2010a-c). Till now, only a little attention has been paid to the sampled-data consensus of the leaderless multi-agent systems with measurement noises (Li and Zhang, 2009a, b) band the sampled-data consensus of the leader-follower multi-agent systems without measurement noises (Hu and Hong, 2007). To the best of our knowledge, there is no open report on the consensus tracking of the leader-follower multi-agent systems with measurement noises based on sampled-data.

Based on the above consideration, this study was concerned with the stochastic bounded consensus tracking problems of leader-follower multi-agent systems. Here, the control input of each agent can only use the information measured at the sampling instants from its neighbors or the virtual leader with a time-varying reference state and the measurements are corrupted by random noises. The probability limit theory and algebra graph theory were employed to derive the sufficient conditions guaranteeing the mean square bounded consensus tracking.

PRELIMINARIES AND PROBLEM STATEMENT

Algebra graph theory: Let G = (V, E, A) be a weighted undirected graph with the set of nodes V = {v1, v2, …, vN}, the set of edges E⊆VxV and the weighted adjacency matrix A = [aij] with nonnegative adjacency elements aij. The node indexes of G belong to a finite index set I = {1, 2, …, N}. An edge of G is denoted by eij = (vi, vj). The adjacency elements associated with the edges are positive, i.e., eijεE⇔aij>0. Moreover, it is assumed that aii = 0 for all iεI. For the undirected graph G, the adjacency matrix A is symmetric, i.e.,aij = aji. The set of neighbors of node vi is denoted by Ni = {vjεV:eijεE}. The degree of node vi is defined as:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data

The Laplacian matrix of G is defined as L = D-A, where D = Diag(d1, d2, …, dN) is the degree matrix of G with diagonal elements di and zero off-diagonal elements. An important fact of L is that all the row sums are zero and thus L has a right eigenvector lN associated with the zero eigenvalue, where lN denotes the N-dimensional column vector with all ones. A path between two distinct nodes vi and vj means a sequence of distinct edges of the form Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data. A graph is called connected if there is a path between any two distinct nodes of the graph. An important property of the Laplacian matrix, which is instrumental in the convergence analysis of consensus protocols, is that the graph G is connected if and only if rank(L) = N-1. Thus, for a connected graph G, L has one and only one zero eigenvalue and all the other eigenvalues of L are positive.

Some notations: The following notations are used throughout this paper. 0N denotes the N-dimensional column vector with all zeros. IN and 0NxN denote the N-dimensional identity matrix and the N-dimensional matrix with all zeros, respectively. For a given vector or matrix M, MT and ||M||2 denote its transpose and 2-norm, respectively. Let R denote the set of real numbers. For a given matrix MεRNxN, det(M) and λi(M) denote its determinant and N eigenvalues. For a given random variable X, EX denotes its mathematical expectation.

Problem statement: In a multi-agent system with N agents, an agent and an available information flow between two agents are considered as a node and an edge in an undirected graph, respectively. Consider the system of dynamic agents described by:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(1)

where, xi(t)εR and ui(t)εR represent, respectively, the state of the ith agent and the associated control input.

The ith agent can receive the information yij(t) = xj(t)+σijωij(t), jεNi from its neighbors, where yij(t) denotes the measurement of the jth agent’s state xj(t) by the ith agent, {ωij(t):jεNi, iεN} are the independent standard Gaussian white noises and σij>0 represents the intensity of the measurement noise ωij(t). If the ith agent has the access to the virtual leader with a time-varying reference state x*(t), then it can receive the information from the virtual leader: yi*(t) = xi*(t)+ρiηi(t), where yi*(t) denotes the measurement of the virtual leader’s state xi*(t) by the ith agent, {ηi(t): iεI} are the independent standard Gaussian white noises and ρi>0 represents the intensity of the measurement noise ηi(t). In this study, it is assumed that the measurement noises {ωij(t):jεNi, iεI} and {ηi(t):iεI} are mutually independent.

In Ma et al. (2008), the following continuous-time consensus tracking protocol was introduced:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(2)

where, a(.):[0, +∞)→[0, +∞) is a piecewise continuous function, called a time-varying consensus gain and the constant feedback gain αi is defined as:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(3)

The protocol Eq. 2 is called a mean square consensus tracking protocol, if the system Eq. 1 using Eq. 2 has the following property: Limt→∞E(xi(t)-x*(t))2 = 0, ∀iεI.

Using the period sampling technology and zero-order hold circuit, the following consensus tracking protocol with sampled data is induced from the continuous-time consensus tracking protocol (2):

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(4)

The discrete-time model with zero-order holder of the system Eq. 1 is:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(5)

which can be rewritten in the vector form:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(6)

where, x(kh) = (x1(kh), x2(kh)…, xN(kh))T.

Definition 1: The protocol (4) is called a mean square bounded consensus tracking protocol, if the system (5) using (4) has the following property:

Limk→∞E(xi(kh)-x*(kh)2 = C<∞, ∀iεI

where, C is a bounded positive constant independent of k.

In the following sections, the sufficient conditions guaranteeing the system (5) using the protocol (4) to achieve the mean square bounded consensus tracking will be derived. Before moving on, the following lemma is needed:

Lemma 1 (Hu and Hong, 2007): If the network topology composed of N agents is fixed, undirected and connected and at least one agent has the access to the virtual leader, then the matrix Diag(α1, α2, …, αN)+L must be positive definite, i.e.:

MiniεI{λi(Diag(α1, α2, …, αN)+L)}>0

In order to better focus on the effects of the sampled-data and measurement noises on the consensus tracking, throughout this paper it is assumed that the communication network topology composed of N agents is fixed, undirected and connected and at least one agent has the access to the virtual leader.

CONVERGENCE ANALYSIS

In this section, the probability limit theory and algebra graph theory were employed to derive the sufficient conditions that guarantee the multi-agent system (5) applying the consensus tracking protocol (4) to reach the mean square bounded consensus tracking.

Theorem 1: Assume that the time-varying reference state of the virtual leader satisfies:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data

Then the multi-agent system (5), applying the consensus tracking protocol (4), achieves the mean square bounded consensus tracking, provided that the following two conditions are satisfied:

C1: Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
C2: Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data

Moreover, if the time-varying reference state of the virtual leader further satisfies, then the system (5) under (4) has the following property:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(7)

Proof: Substituting the protocol (4) into (5) leads to:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(8)

Where:

n(kh) = (n1(kh), n2(kh), …, nN(kh))T

and

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data

Set Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data(kh) = xi(kh)-x*(kh), iεI and Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data(kh) = (Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data(kh), Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data(kh)…, Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data(kh))T. Note that L1N = 0, thus:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(9)

where, P(k) = IN-ha(k) (Diag(α1, α2, …, αN)+L) and Δx*(kh) = x*(kh+h)-x*(kh).

Denote:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(10)

From Eq. 9, it gives:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(11)

It follows from Eq. 11 and the conditions (C1) and (C2) that:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(12)

Employing the conditions (C1) and (C2) yields:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(13)

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(14)

Which leads to:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(15)

and

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(16)

Which can be easily proved by using the analysis similar to one in Eq .14.

Combining Eq. 12, 13, 15 and 16 leads to:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(17)

Therefore, under the given conditions (C1) and (C2), the multi-agent system (5), applying the consensus tracking protocol (4), achieves the mean square bounded consensus tracking.

In addition, if the time-varying reference state further satisfies ξ* = o(1), h→0, then from Eq. 17 it follows:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(18)

Here, the following two equations have been used:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(19)

and

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(20)

Remark 1: If the reference state of the virtual leader is constant, i.e., x*(t)≡x*(0), t≥0, then:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data

Thus, under the conditions (C1) and (C2), the results of Theorem 1 naturally hold.

Remark 2: It is obvious that a(t)≡1 satisfies the condition (C1), which implies that the consensus tracking protocol (4) has the inner robustness to measurement noises. Different from most of the existing work about consensus of multi-agent systems with measurement noises, where the conditions:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data

and

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
(21)

are used for the continuous-time system and the discrete-time system, respectively, the condition (C1) used in Theorem 1 can yield:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data

Notice that the condition (21) means Limk→∞a(k) = 0 and thus INFk≥0a(k) = 0. Using the convergence method in Theorem 1 cannot derive the results of Theorem 1 if the condition (C1) is replaced with the condition (Eq. 21). However, the following results can be obtained:

Corollary 1: Assume that the time-varying reference state of the virtual leader satisfies:

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
Fig. 1: Network topology composed of four agents and a virtual leader

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
Fig. 2: The sates of the system (5) using Eq. 4 with h = 0.2, a(k) = a1(k) and Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data

Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data
Fig. 3: The sates of the system (5) using Eq. 4 with h = 0.2, a(k) = a2(k) and Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data

Then the multi-agent system (5), applying the consensus tracking protocol (4), achieves the mean square consensus tracking, provided that the conditions (21) and (C2) are satisfied.

SIMULATIONS

Numerical simulations were to illustrate the effectiveness of the above theoretical results. Consider a multi-agent system composed of four agents with the network topology shown in Fig. 1, where the agent 0 represents the virtual leader and all the weights of edges are 1. The measurement noises are independent white Gaussian sequences with the intensity equal to 0.5. The initial states of the four agents are x1(0) = -3, x2(0) = -4, x3(0) = 3, x4(0) = 4. Two candidate sequences of time-varying consensus gains are a1(k) = (k+1)/(k+2) and a2(k) = (1+k)-2/3. The numerical results are shown in Fig. 2 and 3, respectively.

From Fig. 2 and 3, it can be seen that the system (5) using (4) with h = 0.2, a(k) = a1(k) and Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data can achieve the mean square bounded consensus tracking, whereas the system (5) using (4) with h = 0.2, a(k) = a2(k) and Image for - Consensus Tracking of Stochastic Multi-agent Systems Based on Sampled-data cannot achieve the mean square bounded consensus tracking. Thus, the results of Theorem 1 are numerically testified.

CONCLUSIONS


In this study, the stochastic bounded consensus tracking problems of leader-follower multi-agent systems with sampling-data, measurement noises and a time-varying reference state were investigated. The probability limit theory and algebra graph theory were employed for deriving the sufficient conditions guaranteeing the mean square bounded consensus tracking. It turned out that the sufficient conditions for guaranteeing the mean square bounded consensus tracking quite depend on the sampling period and the time-varying consensus gains. A future work is to extend the results in this paper to the second-order multi-agent systems with switching directed network topology and the sampling delay.

ACKNOWLEDGMENTS


This study was supported by National Natural Science Foundation of China under Grant No. 61203147, Fundamental Research Funds for the Central Universities of China under Grant No. JUSRP111A44 and Humanities and Social Sciences Youth Funds of the Ministry of Education under Grant No. 12YJCZH218.

REFERENCES


  1. Cao, Y. and W. Ren, 2009. Sampled-data formation control under dynamic directed interaction. Proceedings of the American Control Conference, June 10-12, 2009, St. Louis, MO., pp: 5186-5191.
    CrossRef

  2. Cao, Y. and W. Ren, 2010. Multi-vehicle coordination for double-integrator dynamics under fixed undirected/directed interaction in a sampled-data setting. Int. J. Robust Nonlinear Control, 20: 987-1000.
    CrossRef

  3. Cao, Y. and W. Ren, 2010. Sampled-data discrete-time coordination algorithms for double-integrator dynamics under dynamic directed interaction. Int. J. Control, 83: 506-515.
    CrossRef

  4. Cortes, J., 2006. Analysis and design of distributed algorithms for χ-consensus. Proceedings of Conference on Decision and Control, December 13-15, 2006, San Diego, CA., pp: 3363-3368.

  5. Gao, Y. and L. Wang, 2010. Asynchronous consensus of continuous-time multi-agent systems with intermittent measurements. Int. J. Control, 83: 522-562.
    CrossRef

  6. Gao, Y. and L. Wang, 2010. Consensus of multiple dynamic agents with sampled information. IET Control Theory Appl., 4: 945-956.
    CrossRef

  7. Gao, Y. and L. Wang, 2010. Consensus of multiple double-integrator agents with intermittent measurement. Int. J. Robust Nonlinear Control, 20: 1140-1155.
    CrossRef

  8. Gao, Y., L. Wang, G.M. Xie and B. Wu, 2009. Consensus of multi-agent systems based on sampled-data control. Inter. J. Control, 82: 2193-2205.
    CrossRef

  9. Hayakawat, T., T. Matsuzawat and S.J. Hara, 2006. Formation control of multi-agent systems with sampled information: Relationship between information exchange structure and control performance. Proceedings of the 45th IEEE Conference Decision and Control, December 13-15, 2006, San Diego, CA., pp: 4333-4338.
    CrossRef

  10. Hu, J. and G. Feng, 2010. Distributed tracking control of leader-follower multi-agent systems under noisy measurement. Automatica, 46: 1382-1387.
    CrossRef

  11. Hu, J. and Y. Hong, 2007. Leader-following coordination of multi-agent systems with coupling time delays. Physica A: Stat. Mech. Appl., 374: 853-863.
    CrossRef

  12. Huang, M. and J.H. Manton, 2009. Coordination and consensus of networked agents with noisy measurements: Stochastic algorithms and asymptotic behavior. SIAM J. Control Optim., 48: 134-161.
    Direct Link

  13. Huang, M. and J.H. Manton, 2007. Stochastic double array analysis and convergence of consensus algorithms with noisy measurements. Proceedings of the American Control Conference, July 9-13, 2007, New York, pp: 705-710.
    CrossRef

  14. Huang, M. and J.H. Manton, 2007. Stochastic Lyapunov analysis for consensus algorithms with noisy measurements. Proceedings of the American Control Conference, July 9-13, 2007, New York, pp: 1419-1424.

  15. Huang, M. and J.H. Manton, 2010. Stochastic consensus seeking with noisy and directed inter-agent communication: Fixed and randomly varying topologies. IEEE Trans. Autom. Control., 55: 235-241.
    CrossRef

  16. Jadbabaie, A., L. Jie and A.S. Morse, 2003. Coordination of groups of mobile autonomous agents using nearest neighbor rules. IEEE Trans. Automat. Control, 48: 988-1001.
    CrossRefDirect Link

  17. Kingston, D.B., W. Ren and R.W. Beard, 2005. Consensus algorithm are input-to-state stable. Proc. Amer. Control Conf., 3: 1686-1690.
    CrossRef

  18. Li, T. and J.F. Zhang, 2010. Consensus conditions of multi-agent systems with time-varying topologies and stochastic communication noises. IEEE Trans. Autom. Control, 55: 2043-2057.
    CrossRef

  19. Li, T. and J.F. Zhang, 2009. Mean square average consensus under measurement noises and fixed topologies: Necessary and sufficient conditions. Automatica, 45: 1929-1936.
    CrossRef

  20. Li, T. and J.F. Zhang, 2009. Sampled-data based average consensus with measurement noises: Convergence analysis and uncertainty principle. Sci. China Ser. F-Inf. Sci., 52: 2089-2103.
    CrossRef

  21. Ma, C.Q., T. Li and J.F. Zhang, 2008. Leader-following consensus control for multi-agent systems under measurement noises. Proceedings of the 17th World Congress the International Federation of Automatic Control, July 6-11, 2008, Seoul, Korea, pp: 1528-1533.
    CrossRef

  22. Mohanarajah, G. and T. Hayakawa, 2008. Formation stability of multi-agent systems with limited information. Proceedings of the American Control Conference, June 11-13, 2008, Seattle, WA., pp: 704-709.
    CrossRef

  23. Murray, R.M., 2007. Recent research in cooperative control of multi-vehicle systems. J. Dyn. Syst. Meas. Control, 129: 571-583.
    CrossRef

  24. Saber, R.O., J.A. Fax and R.M. Murray, 2007. Consensus and cooperation in multi-agent networked systems. Proc. IEEE, 95: 215-233.
    CrossRefDirect Link

  25. Olfati-Saber, R. and R.M. Murray, 2004. Consensus problems in networks of agents with switching topology and time-delays. IEEE Trans. Autom. Control, 49: 1520-1533.
    CrossRefDirect Link

  26. Ren, W. and R.W. Beard, 2005. Consensus seeking in multiagent systems under dynamically changing interaction topologies. IEEE Trans. Automat. Control, 50: 655-661.
    CrossRefDirect Link

  27. Ren, W., R.W. Beard and E.M. Atkins, 2007. Information consensus in multivehicle cooperative control. IEEE Control Syst. Mag., 27: 71-82.
    CrossRef

  28. Ren, W., R.W. Beard and D.B. Kingston, 2005. Multi-agent Kalman consensus with relative uncertainty. Proceedings of the American Control Conference, Jun 8-10, 2005, IEEE Computer Society, pp: 1865-1870.

  29. Ren, W. and Y.C. Cao, 2008. Convergence of sampled-data consensus algorithms for double-integrator dynamics. Proceedings of the 47th IEEE Conference on Decision and Control, December 9-11, 2008, Cancun, pp: 3965-3970.

  30. Xiao, L., S. Boyd and S.J. Kim, 2007. Distributed average consensus with least-mean square deviation. J. Parallel Distrib. Comput., 67: 33-46.

Leave a Comment


Your email address will not be published. Required fields are marked *

Useful Links

  • Journals
  • For Authors
  • For Referees
  • For Librarian
  • For Socities

Contact Us

Office Number 1128,
Tamani Arts Building,
Business Bay,
Deira, Dubai, UAE

Phone: +971 507 888 742
Email: [email protected]

About Science Alert

Science Alert is a technology platform and service provider for scholarly publishers, helping them to publish and distribute their content online. We provide a range of services, including hosting, design, and digital marketing, as well as analytics and other tools to help publishers understand their audience and optimize their content. Science Alert works with a wide variety of publishers, including academic societies, universities, and commercial publishers.

Follow Us
© Copyright Science Alert. All Rights Reserved