Avalanche Victim Search via Robust Observers

This article deals with the problem of victim localization in avalanches by using controlled unmanned aerial vehicles (UAVs) equipped with an electromagnetic sensor (known as ARVA) typically adopted in these search and rescue scenarios. We show that the nominal ARVA measurement can be linearly related to a quantity that is sufficient to reconstruct the victim position. We explicitly deal with a robust scenario in which the measurement is actually perturbed by the noise that grows with the distance to the victim and propose an adaptive control scheme based on a least-square identifier and a trajectory generator whose role is both to guarantee the persistence of excitation for the identifier and to steer the ARVA receiver toward the victim. We prove that the controller ensures boundedness of trajectories and enables to localize the victim in a domain where the ARVA output is sufficiently informative. We illustrate its performance in a realistic simulation framework specifically developed with real data. The proposed approach could significantly reduce the searching time by providing an exploitable estimate before having reached the victim.