Alouette van Hove, Kristoffer Aalstad, Norbert Pirk
Published 2024-01-08Version 1
The accurate estimation of locations and emission rates of gas sources is crucial across various domains, including environmental monitoring and greenhouse gas emission analysis. This study investigates two drone sampling strategies for inferring source term parameters of gas plumes from atmospheric measurements. Both strategies are guided by the goal of maximizing information gain attained from observations at sequential locations. Our research compares the myopic approach of infotaxis to a far-sighted navigation strategy trained through deep reinforcement learning. We demonstrate the superior performance of deep reinforcement learning over infotaxis in environments with non-isotropic gas plumes.
Comments: To be published in Proceedings of Machine Learning Research (Proceedings of the 5th Northern Lights Deep Learning Conference (NLDL))
Journal: Proceedings of Machine Learning Research (PMLR) 233 (2024) 89-96
Playing Atari with Deep Reinforcement Learning
Communication Topologies Between Learning Agents in Deep Reinforcement Learning
Massively Parallel Methods for Deep Reinforcement Learning
![QR code for arXiv:2401.03947 [cs.LG]](http://oss.arxitics.com/images/favicon.svg)