Evaluating Alternative Solutions for Aircraft Collision Avoidance

Abstract

MS Thesis Defense:

The proliferation of small unmanned aircraft systems (sUAS) has caused modern airspace regulation authorities to examine the interoperability of these aircraft with collision avoidance systems initially designed for large commercial aircraft. Limitations in the currently mandated system led the Federal Aviation Administration to commission the development of a new solution, the Airborne Collision Avoidance System X (ACAS X), designed to enable a collision avoidance capability for multiple aircraft platforms, including sUAS.

This work explores the assessment of ACAS X by analyzing alternative solutions. We establish a qualitative classification scheme for collision avoidance strategies, implement an algorithm that prioritizes operational considerations unique to sUAS, and compare it to the leading approach. We then tune an existing deep reinforcement learning algorithm’s parameters with a surrogate optimizer. Our experiments show that these optimized parameters increase safety and operational viability and support future capability development for sUAS collision avoidance.

Join meeting in Cummings 074.