Resilient Randomized Quantized Consensus

Dibaji, Seyed Mehran; Ishii, Hideaki; Tempo, Roberto

doi:10.1109/tac.2017.2771363

Cited by 133 publications

(79 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, (7) shows that the state of the compromised agent j, i.e., x j , is deviated from the desired consensus value with a value proportional to f j . Therefore, (27) results in deviating the state of the immediate neighbor of the compromised agent j from the desired consensus behavior, which contradicts the assumption. Consequently, intact agents that have a path to the compromised agent do not reach consensus, while their local neighborhood tracking error is zero.…”

Section: B Attack Analysismentioning

confidence: 93%

Resilient synchronization of distributed multi-agent systems under attacks

2020

View full text Add to dashboard Cite

In this paper, we first address adverse effects of cyber-physical attacks on distributed synchronization of multiagent systems, by providing conditions under which an attacker can destabilize the underlying network, as well as another set of conditions under which local neighborhood tracking errors of intact agents converge to zero. Based on this analysis, we propose a Kullback-Liebler divergence based criterion in view of which each agent detects its neighbors' misbehavior and, consequently, forms a self-belief about the trustworthiness of the information it receives. Agents continuously update their selfbeliefs and communicate them with their neighbors to inform them of the significance of their outgoing information. Moreover, if the self-belief of an agent is low, it forms trust on its neighbors. Agents incorporate their neighbors' self-beliefs and their own trust values on their control protocols to slow down and mitigate attacks. We show that using the proposed resilient approach, an agent discards the information it receives from a neighbor only if its neighbor is compromised, and not solely based on the discrepancy among neighbors' information, which might be caused by legitimate changes, and not attacks. The proposed approach is guaranteed to work under mild connectivity assumptions.

show abstract

Section: B Attack Analysismentioning

confidence: 93%

Resilient synchronization of distributed multi-agent systems under attacks

2020

View full text Add to dashboard Cite

show abstract

“…By having n a i,k as a weight, establishing a connection with a neighbor is penalized more if the number of attacks detected increases. Moreover, (25b) is obtained from (3)-(7), (13), (14) and from the following expression:…”

Section: Figurementioning

confidence: 99%

“…Moreover, consensus problems in which some of the agents perform an adversarial behavior to prevent convergence have also been investigated. [13][14][15] In the DMPC framework, the issue that some agents might provide false information, which is a different type of adversarial behaviors, has also been discussed. 16,17 A scenario-based defense mechanism 16 and a compensation scheme to incentivize truth telling among agents 17 have been proposed to deal with false information problems.…”

Section: Introductionmentioning

confidence: 99%

Resilient distributed model predictive control for energy management of interconnected microgrids

Ananduta

Maestre

Ocampo‐Martínez

et al. 2019

Optim Control Appl Methods

Self Cite

View full text Add to dashboard Cite

Summary Distributed energy management of interconnected microgrids that is based on model predictive control (MPC) relies on the cooperation of all agents (microgrids). This paper discusses the case in which some of the agents might perform one type of adversarial actions (attacks) and they do not comply with the decisions computed by performing a distributed MPC algorithm. In this regard, these agents could obtain a better performance at the cost of degrading the performance of the network as a whole. A resilient distributed method that can deal with such issues is proposed in this paper. The method consists of two parts. The first part is to ensure that the decisions obtained from the algorithm are robustly feasible against most of the attacks with high confidence. In this part, we formulate the economic dispatch problem, taking into account the attacks as a chance‐constrained problem, and employ a two‐step randomization‐based approach to obtain a feasible solution with a predefined level of confidence. The second part consists in the identification and mitigation of the adversarial agents, which utilizes hypothesis testing with Bayesian inference and requires each agent to solve a mixed‐integer problem to decide the connections with its neighbors. In addition, an analysis of the decisions computed using the stochastic approach and the outcome of the identification and mitigation method is provided. The performance of the proposed approach is also shown through numerical simulations.

show abstract

“…Recent works Koutsoukos, 2011, 2012) proposed a continuoustime variation of the MSR algorithm, while in (Su and Vaidya, 2017) the authors allowed agents to take as input not only the information from their neighbors but also the information provided by the agents that are up to hops away. In (Dibaji et al, 2018;Dibaji and Ishii, 2017), a modification of the MSR algorithm has been proposed to solve the resilient consensus problem for asynchronous multi-agent systems with quantized communication and delays. The works (Sundaram and Hadjicostis, 2011;Pasqualetti et al, 2012) proposed a different approach to detect and identify malicious agents, which is based on observability and reconstruction of the initial conditions.…”

Section: Introductionmentioning

confidence: 99%

Resilient consensus for multi-agent systems subject to differential privacy requirements

Fiore

Russo

2019

Automatica

View full text Add to dashboard Cite

We consider multi-agent systems interacting over directed network topologies where a subset of agents is adversary/faulty and where the non-faulty agents have the goal of reaching consensus, while fulfilling a differential privacy requirement on their initial conditions. To address this problem, we develop an update law for the non-faulty agents. Specifically, we propose a modification of the so-called Mean-Subsequence-Reduced (MSR) algorithm, the Differentially Private MSR (DP-MSR) algorithm, and characterize three important properties of the algorithm: correctness, accuracy and differential privacy. We show that if the network topology is (2f + 1)-robust, then the algorithm allows the non-faulty agents to reach consensus despite the presence of up to f faulty agents and we characterize the accuracy of the algorithm. Furthermore, we also show in two important cases that our distributed algorithm can be tuned to guarantees differential privacy of the initial conditions and the differential privacy requirement is related to the maximum network degree. The results are illustrated via simulations.

show abstract

Resilient Randomized Quantized Consensus

Cited by 133 publications

References 41 publications

Resilient synchronization of distributed multi-agent systems under attacks

Resilient synchronization of distributed multi-agent systems under attacks

Resilient distributed model predictive control for energy management of interconnected microgrids

Resilient consensus for multi-agent systems subject to differential privacy requirements

Contact Info

Product

Resources

About