Expectation-Maximization Based Defense Mechanism for Distributed Model Predictive Control

Rafael Accácio Nogueira (1, 2) , Romain Bourdais (2, 1) , Simon Leglaive (2, 1) , Hervé Guéguen (2, 1)

1 CentraleSupélec [campus de Rennes]
2 IETR - Institut d'Électronique et des Technologies du numéRique

Controlling large-scale systems sometimes requires decentralized computation. Communication among agents is crucial to achieving consensus and optimal global behavior. These negotiation mechanisms are sensitive to attacks on those exchanges. This paper proposes an algorithm based on Expectation Maximization to mitigate the effects of attacks in a resource allocation based distributed model predictive control. The performance is assessed through an academic example of the temperature control of multiple rooms under input power constraints.

Keywords: Distributed constrained control and MPC; Cyber security in networked control systems; Decentralized Control and Large-Scale Systems

Domains: Automatic; Cryptography and Security [cs.CR]

Rafael Accácio Nogueira, Romain Bourdais, Simon Leglaive, Hervé Guéguen. Expectation-Maximization Based Defense Mechanism for Distributed Model Predictive Control. 9th IFAC Conference on Networked Systems (NecSys22), Jul 2022, Zürich, Switzerland. ⟨hal-03723298⟩