Drifters are free-floating measuring devices of ocean currents that are being increasingly deployed by oceanographers. Such devices, once placed, will flow according to the currents and collect data accordingly. They are favoured by the practitioners due to their relatively low costs and ability to capture both the spatial and temporal characteristics of the ocean currents. The deployment strategy of such sensors, however, is understudied, with the majority of existing placement campaigns being either following standard 'space-filling' designs or relatively ad-hoc expert opinions. A key challenge to applying principled active learning in this setting is that Lagrangian observers are continuously advected through the vector field, so they make measurements at different locations and times. It is, therefore, important to consider the likely future trajectories of placed observers to account for the utility of candidate placement locations. To this end, we propose a formal active learning framework for drifter placement that accounts for the structure of drifter observations and present BALLAST: Bayesian Active Learning with Look-ahead Amendment for Sea-drifter Trajectories. We observe noticeable benefits of BALLAST-aided sequential observer placement strategies on both synthetic and high-fidelity ocean current models.
