UBAT: On Jointly Optimizing UAV Trajectories and Placement of Battery Swap Stations

TL;DR

This paper introduces UBAT, a heuristic framework using ant colony optimization to jointly optimize UAV trajectories and the placement of charging stations, significantly improving deployment efficiency.

Contribution

The paper presents a novel heuristic approach for the NP-hard problem of deploying charging stations and optimizing UAV trajectories simultaneously.

Findings

UBAT achieves solutions within 8.3% of optimal for UAV trajectories.

UBAT achieves solutions within 7.3% of optimal for charging station placement.

Extensive simulations validate the effectiveness of UBAT in multi-objective optimization.

Abstract

Unmanned aerial vehicles (UAVs) have been widely used in many applications. The limited flight time of UAVs, however, still remains as a major challenge. Although numerous approaches have been developed to recharge the battery of UAVs effectively, little is known about optimal methodologies to deploy charging stations. In this paper, we address the charging station deployment problem with an aim to find the optimal number and locations of charging stations such that the system performance is maximized. We show that the problem is NP-Hard and propose UBAT, a heuristic framework based on the ant colony optimization (ACO) to solve the problem. Additionally, a suite of algorithms are designed to enhance the execution time and the quality of the solutions for UBAT. Through extensive simulations, we demonstrate that UBAT effectively performs multi-objective optimization of generation of UAV trajectories and placement of charging stations that are within 8.3% and 7.3% of the true optimal solutions, respectively.