Secure state estimation for Cyber Physical Systems with sparse malicious packet drops

In this paper we investigate how to correctly estimate the state of a linear system in a discrete time setting, when the communication between sensors and/or actuators with the controller is corrupted by sparse attacks. We give conditions for the system to be resilient against malicious packet drops without assuming that the attacker follows any specific probabilistic model. The performance of the proposed algorithm is demonstrated through numerical simulations, where a flexible payload is transported by a group of Unmanned Aerial Vehicles (UAVs).