Environmental Hotspot Identification in Limited Time with a UAV Equipped with a Downward-Facing Camera

TL;DR

This paper introduces a novel multi-fidelity Gaussian Process bandit approach for UAV-based environmental hotspot detection, leveraging correlated rewards and image-based measurements to efficiently identify maxima within limited flight time.

Contribution

It formulates a new multi-fidelity bandit problem incorporating correlated rewards and image quality dependent on UAV altitude, with an empirical sensing strategy.

Findings

Proposed method effectively finds hotspots with limited UAV flight time.

Empirical results show improved accuracy over baseline strategies.

Scalability demonstrated in large-scale simulated environments.

Abstract

Our work is motivated by environmental monitoring tasks, where finding the global maxima (i.e., hotspot) of a spatially varying field is crucial. We investigate the problem of identifying the hotspot for fields that can be sensed using an Unmanned Aerial Vehicle (UAV) equipped with a downward-facing camera. The UAV has a limited time budget which it can use for learning the unknown field and identifying the hotspot. Our contribution is to show how this problem can be formulated as a novel multi-fidelity variant of the Gaussian Process (GP) multi-armed bandit problem. The novelty is two-fold: (i) unlike standard multi-armed bandit settings, the rewards of the arms are correlated with each other; and (ii) unlike standard GP regression, the measurements in our problem are images (i.e., vector measurements) whose quality depends on the altitude of the UAV. We present a strategy for finding the sequence of UAV sensing locations and empirically compare it with several baselines. Experimental results using images gathered onboard a UAV are also presented and the scalability of the proposed methodology is assessed in a large-scale simulated environment in Gazebo.