Numerical investigation of the spatial estimation error in sensor guidance used for the localization of a gaseous source in a 2D domain

An undesirable release of a gaseous substance in the atmosphere can pose a significant threat. To minimize the negative impact of this release, it is important to predict the state of the release as well as the proximity of the source location. This work investigates the real time detection and state estimation of an aerial release due to a stationary or moving source in a 2D domain. A mobile sensing agent is used to provide state measurements at desired locations using a Lyapunov based guidance scheme. Several computational results illustrate the success of the sensor guidance scheme. State reduction and state dependent grid adaptation are utilized to increase computational efficiency and to provide real time state estimation results.