Modelling and control of a flying robot interacting with the environment

This work focuses on the problem of modelling and controlling a Ducted-Fan Miniature unmanned Aerial Vehicle (DFMAV) considering explicitly the interaction with the environment and the resulting constraints that affect the system dynamics. The goal is to address a scenario in which DFMAVs accomplish tasks requiring contact between the aerial vehicle and the environment such as remote manipulation, docking and flight in cluttered environments. Since the system’s dynamics may be dramatically different when contacts happen and when they do not, an overall description of the system is obtained by collection of the different behaviours into a hybrid automaton. For this particular class of hybrid dynamical systems, a framework for robust control of the system based on a path following strategy is developed and tested on a scenario in which the DFMAV is required to dock with and undock from a vertical surface. Simulation results are also presented to show the effectiveness of the proposed framework.