Abstract
Evolution Strategies (ESs) have been successfully applied to dynamic system identification; that is, to tuning the parameters of (models of) dynamic systems. At the Renato Archer Research Center (Cenpra), Project AURORA is developing a semi-autonomous lighter-than-air craft for aerial inspection and environmental monitoring. The development of an autonomous control system for the target airship requires an accurate dynamic model of it. An ES has been used to identify values for the parameters of such a model, and some simulation results are presented here. The approach proposed has the final objective of being applied to real flight data to identify the parameters of the airship itself.