Routing Problem for Unmanned Aerial Vehicle Patrolling Missions -- A Progressive Hedging Algorithm

TL;DR

This paper develops a two-stage stochastic model and a progressive hedging algorithm to optimize UAV routing for patrolling missions, accounting for information fidelity and supplemental target visits.

Contribution

It introduces a novel stochastic programming approach with a progressive hedging algorithm for UAV routing with information quality considerations.

Findings

The proposed model effectively captures the decision-making process under uncertainty.

The progressive hedging algorithm provides high-quality solutions efficiently.

Computational results demonstrate the model's practical applicability and robustness.

Abstract

The paper presents a two-stage stochastic program to model a routing problem involving an Unmanned Aerial Vehicle (UAV) in the context of patrolling missions. In particular, given a set of targets and a set of supplemental targets corresponding to each target, the first stage decisions involve finding the sequence in which the vehicle has to visit the set of targets. Upon reaching each target, the UAV collects information and if the operator of the UAV deems that the information collected is not of sufficient fidelity, then the UAV has to visit all the supplemental targets corresponding to that target to collect additional information before proceeding to visit the next target. The problem is solved using a progressive hedging algorithm and extensive computational results corroborating the effectiveness of the proposed model and the solution methodology is presented.