A controls-CFD approach for estimation of concentration from a moving aerial source: Comparisons between switched and dynamically adapted grids in 2D

Examining control and estimation algorithms from the computational perspective, this work proposes a dynamic grid adaptation scheme for the concentration estimation of a mobile gaseous source. The guidance of a moving sensor on board a sensing aerial vehicle, is solely dictated by the performance of the estimation scheme. To further improve the estimator performance, a dynamic grid adaptation scheme is proposed and which adapts the grid via local refinement and coarsening by using the current position of the sensor. This essentially renders the state matrix of the state estimator time varying and this time variation is dictated by the sensor spatial location. Extensive numerical studies for both stationary and mobile sources are considered to provide a convincing argument on the performance-based guidance of mobile sensors.