Algorithms for persistent autonomy and surveillance

Abstract

In this thesis, we consider the problem of monitoring stochastic, time-varying events occurring at discrete locations. Our problem formulation extends prior work in persistent surveillance by considering the objective of successfully completing monitoring tasks in unknown, dynamic environments where the rates of events are time-inhomogeneous and may be subject to abrupt changes. We propose novel monitoring algorithms that effectively strike a balance between exploration and exploitation as well as a balance between remembering and discarding information to handle temporal variations in unknown environments. We present analysis proving the favorable properties of the policies generated by our algorithms and present simulation results demonstrating their effectiveness in several monitoring scenarios inspired by real-world applications. Our theoretical and empirical results support the applicability of our algorithm to a wide range of monitoring applications, such as detection and tracking efforts at a large scale.