Abstract:
This paper presents an attack-resilient estimation scheme for uniformly observable nonlinear systems having redundant sensors when a subset of sensors is corrupted by adv...Show MoreMetadata
Abstract:
This paper presents an attack-resilient estimation scheme for uniformly observable nonlinear systems having redundant sensors when a subset of sensors is corrupted by adversaries. We first design an individual high-gain observer from each measurement output so that partial information of system state is obtained. Then, a nonlinear error correcting problem is formulated by collecting all the information from each partial observers and it can be solved by exploiting redundancy. For most of the time, a computationally efficient monitoring system is running and it detects every influential attacks. A simple switching logic explores another combination of output measurement to find a correct candidate only when a residual signal exceeds its threshold.
Published in: 2016 IEEE 55th Conference on Decision and Control (CDC)
Date of Conference: 12-14 December 2016
Date Added to IEEE Xplore: 29 December 2016
ISBN Information: