Resilient Distributed State Estimation and Hypothesis Testing in Dynamic Environments
Primary Investigator:
Shreyas Sundaram
Aritra Mitra, John A. Richards, Saurabh Bagchi and Shreyas Sundaram
Abstract
Applications in environmental monitoring, surveillance, and industrial control require a network of (mobile) agents to cooperatively understand the state of the world based on their local measurements and the information that they obtain from their neighbors. However, these networks of mobile agents also introduce various challenges for distributed state estimation and hypothesis testing, including intermittent or partially informative observations of the system state, loss of communication links due to mobility and packet drops, and the potential for misbehavior by some of the agents. In this work, we describe new ways to solve distributed state estimation and hypothesis testing problems that account for these various challenges. We first propose a simple method to design distributed observers for dynamical systems for the most general class of systems and networks possible, and show how such observers can be extended to handle a broad class of time-varying networks. We then tackle the problem of distributed hypothesis testing, and propose a novel technique to allow each node in the network to rapidly converge to the true state of the world. We further describe how our approaches can be extended to be resilient to adversarial nodes that inject misinformation into the network, or do not follow the prescribed algorithms.