Gas Source Localization Using physics Guided Neural Networks
This addresses gas source localization for robotics and environmental monitoring, presenting an incremental improvement by avoiding costly numerical simulations.
The paper tackles the problem of gas source localization using spatially distributed concentration measurements, proposing a Physics-Guided Neural Network to approximate gas dispersion and efficiently solve the inverse problem, with experiments showing effective localization even under noisy conditions.
This work discusses a novel method for estimating the location of a gas source based on spatially distributed concentration measurements taken, e.g., by a mobile robot or flying platform that follows a predefined trajectory to collect samples. The proposed approach uses a Physics-Guided Neural Network to approximate the gas dispersion with the source location as an additional network input. After an initial offline training phase, the neural network can be used to efficiently solve the inverse problem of localizing the gas source based on measurements. The proposed approach allows avoiding rather costly numerical simulations of gas physics needed for solving inverse problems. Our experiments show that the method localizes the source well, even when dealing with measurements affected by noise.