CONFLICT AVOIDANCE SOLUTIONS FOR CIVIL UNMANNED AIRCRAFT

This thesis presents a study of a possible solution to promote and support the integration of unmanned aircraft in civil non-segregated airspaces. The capability to "detect and avoid" mid-air collision hazards is one of the main technical challenges restricting widespread operations of unmanned aircraft in non-segregated airspaces. In fact, to operate in prescribed environments, an unmanned aircraft needs an onboard technology to replace the capability of the human pilot to "see and avoid" potential collisions with an equivalent level of safety. Such a technology is a so-called "detect and avoid" system. Within the problem of developing such a "detect and avoid" system, this thesis focuses on the corresponding "avoid function" and proposes a possible solution. The approach to obtain the solution is to schematize a generic obstacle or a moving intruder through a quite generic moving ellipsoid that represents the region of space the unmanned aircraft must not violate. The outcome is a novel and deterministic solution that enables situations of potential conflict to be detected through a single metric, both simple and flexible, and to be avoided through a set of possible actions, such as speed changes in magnitude and/or direction. A simulation campaign is also conducted to validate the proposed solution. The result is that the corresponding algorithms are able to implement a process of conflict detection and resolution, in a three-dimensional space and in real-time, even without the assistance of the human operator. Moreover, simulations with the human operator in the loop demonstrate that the developed solution is able to work as an advisory system that provides also resolution indications for the remote human pilot, as well as an autonomous "detect and avoid" system that detects the potential conflict and implements actions to avoid it. As such, the proposed solution can be considered as a fundamental step for the development of a prototype of "detect and avoid" system for promoting the integration of unmanned aircraft in non-segregated civil airspaces.