Persistent Surveillance of Stochastic Events with Unknown Statistics
نویسنده
چکیده
We consider the use of a mobile agent to monitor stochastic, transient events that occur in discrete locations in the environment with the objective of maximizing the number of event observations in a balanced manner. We assume that the events of interest at each station follow a stochastic process with an initially unknown and station-specific rate parameter; Consequently, we are faced with a bandit problem -that is similar to the canonical Multi-Armed Bandit roblemin which the inherent trade-off between exploration and exploitation must be balanced in an appropriate manner. We present a novel monitoring algorithm with provable guarantees that leverages variance estimates to generate policies capable of simultaneously taking into account the pertinent monitoring objectives and the balance between exploration and exploitation. Figure 1: A persistent monitoring application in which a documentary maker would like to monitor three different species of birds is shown above. At each discrete station i, the sightings of birds (i.e. events) follow a stochastic process with rate λi that is initially unknown to the documentary maker and must be learned and approximated throughout the monitoring process. Given a cyclic path defining the sequence of stations to visit, the documentary maker would like to traverse this cyclic path repeatedly, stopping at each station for some defined amount of time in order to observe events. The overarching goal of the documentary maker is to generate and execute a policy π := (t1, t2, t3) that allows the documentary maker to collect the maximum number of bird sightings in a balanced manner across the different species so that no particular bird species receives too little or too much attention.
منابع مشابه
Persistent Surveillance of Events with Unknown Rate Statistics
We present a novel algorithm for persistent monitoring of stochastic events that occur at discrete locations in the environment with unknown event rates. Prior research on persistent monitoring assumes knowledge of event rates, which is often not the case in robotics applications. We consider the multi-objective optimization of maximizing the total number of events observed in a balanced manner...
متن کاملAlgorithms for Persistent Autonomy and Surveillance
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 timeinhomogeneous and may be subject to abrupt changes. We propose novel...
متن کاملThe Need to Update Respiratory Disease Surveillance System and Indigenization of the Pneumonia of Unknown Cause Surveillance System in Iran according to the COVID-19 Pandemic
The Need to Update Respiratory Disease Surveillance System and Indigenization of the Pneumonia of Unknown Cause Surveillance System in Iran according to the COVID-19 Pandemic Seyed Hasan Adeli1, Mohammad Aghaali2 1 Associate Professor, Department of Internal Medicine, Faculty of Medicine, Qom University of Medical Sciences, Qom, Iran 2 Assistant Professor, Department of ...
متن کاملDynamic Inventory Planning with Unknown Costs and Stochastic Demand
Generally ordering policies are done by two methods, including fix order quantity (FOQ) and fix order period (FOP). These methods are static and either the quantity of ordering or the procedure of ordering is fixing in throughout time horizon. In real environments, demand is varying in any period and may be considered as uncertainty. When demand is variable in any period, the traditional and st...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2016