NextGen UAVs: Modeling, Navigation and Control Challenges, and Integration into the National Airspace

Abstract

Research and development in Unmanned Aerial Vehicles (UAVs), or Unmanned Aircraft Systems (UAS), have witnessed unprecedented growth in a wide spectrum of military, civil/public domain applications. Societal impact is major and as such, despite recent technology advancements and breakthroughs, ‘robot ethics’ and ‘ethical AI’ are becoming centerstage issues that need to be addressed, too. The purpose of this seminar is twofold: After offering a short ‘global perspective’ of the current UAV/UAS landscape, it first introduces NextGen platforms that include hybrid, reconfigurable and bioinspired designs, morphing and Circulation Control Wing (CCW) designs, as well as multi-mode (aerial, ground, sea-surface, underwater) ensembles. Challenges related to modeling, navigation and controller design are addressed, and a comprehensive framework comprising model-based and data-driven approaches coupled with a learning-based foundation is detailed, implemented and tested on several aerial platforms. Moreover, the true and complete equivalence of the Newton-Euler (N-E) and Euler-Lagrange (E-L) modeling formulations as applied to multirotor UAVs is proven by deriving a corrected Euler-Lagrange (c-L-E) multirotor attitude dynamics model, as review of the published literature reveals that the commonly adopted E-L multirotor dynamics model is equivalent to the N-E model only when it comes to position dynamics, but not in attitude dynamics. A benchmark framework is derived for controller implementation and comparison via simulations, simulated experiments and actual flight tests. Secord, integration of unmanned aviation into the national airspace is addressed and equivalent levels of safety are introduced to facilitate manned – unmanned functionality in congested airspaces. Several case studies are presented not only to support the discussed technical approaches, but to also pave the way to new research directions with the aim to build highly confident autonomous robot systems.