Unmanned Aircraft Systems challenges in design for autonomy

Unmanned Aircraft Systems, UAS, have seen unprecedented levels of growth during the last decade. Projections and expectations for future UAS utilization span a very wide and diverse spectrum of civilian and public domain applications, in addition to the obvious military applications, from emergency response, to environmental monitoring, early fire detection and forest protection, to name but a few such applications. However, before timely and orderly integration into the national airspace system, NAS, it is essential that challenges at least in the areas of design for autonomy, navigation, robust and fault-tolerant control, sense-detect-and-avoid/see-and-avoid systems for mid-air collision avoidance, UAV safety and reliability, reaches maturity before complete UAS integration into the national airspace system occurs. This plenary contribution first discusses the design for autonomy challenge and the transition from the `human-in-the-loop' to the `human-on-the-loop' concept that is coupled with the much needed reduced operator workload, followed by a comprehensive and modular UAS control architecture aiming at facilitating software developments regardless of specific hardware. A generalized, sensor-based, fault-tolerant navigation control architectural framework for (nonlinear, linearized and linear) autonomous UAS, including a methodology to accommodate in real-time rotorcraft main/tail rotor failures (resulting in helicopter safe landing) is also recommended; however, this framework is also suitable for other types of UAS, i.e., fixed-wing aircraft and multi-rotor configurations.

I. Introduction

This plenary contribution summarizes research findings by the author's group in the area of sensor-based navigation and control of autonomous UAS. The long-term goal of the research is to design, develop and test high-confidence unmanned systems endowed with assured autonomy and ability of autonomous functionality under nominal (error-free) and single- or multiple- point fault conditions.