Efficiently Identifying Hotspots in a Spatially Varying Field with Multiple Robots

TL;DR

This paper introduces adaptive algorithms for environmental hotspot detection using single and multiple robots, employing decentralized coordination and path planning, with findings indicating hyperparameter estimation may be unnecessary.

Contribution

It presents a flexible, hyperparameter-agnostic approach for multi-robot environmental hotspot detection using Voronoi partitioning and Monte Carlo Tree Search.

Findings

Hyperparameter estimation may be unnecessary for hotspot detection.

Multi-robot coordination improves coverage and efficiency.

Algorithms perform well on synthetic and real-world data.

Abstract

In this paper, we present algorithms to identify environmental hotspots using mobile sensors. We examine two approaches: one involving a single robot and another using multiple robots coordinated through a decentralized robot system. We introduce an adaptive algorithm that does not require precise knowledge of Gaussian Processes (GPs) hyperparameters, making the modeling process more flexible. The robots operate for a pre-defined time in the environment. The multi-robot system uses Voronoi partitioning to divide tasks and a Monte Carlo Tree Search for optimal path planning. Our tests on synthetic and a real-world dataset of Chlorophyll density from a Pacific Ocean sub-region suggest that accurate estimation of GP hyperparameters may not be essential for hotspot detection, potentially simplifying environmental monitoring tasks.