Article
Method to Characterize Potential UAS Encounters
Using Open Source Data
Andrew Weinert
MIT Lincoln Laboratory, Lexington, MA 02420-9176, USA; andrew.weinert@ll.mit.edu
Received: 6 October 2020; Accepted: 31 October 2020; Published: 4 November 2020
Abstract:
As unmanned aerial systems (UASs) increasingly integrate into the US national airspace
system, there is an increasing need to characterize how commercial and recreational UASs may
encounter each other. To inform the development and evaluation of safety critical technologies,
we demonstrate a methodology to analytically calculate all potential relative geometries between
different UAS operations performing inspection missions. This method is based on a previously
demonstrated technique that leverages open source geospatial information to generate representative
unmanned aircraft trajectories. Using open source data and parallel processing techniques, we performed
trillions of calculations to estimate the relative horizontal distance between geospatial points across
sixteen locations.
Keywords:
unmanned aerial vehicles; drones; aerospace control; simulation; geospatial analysis;
open source software
1. Introduction
The continuing integration of unmanned aerial system (UAS) operations into the National Airspace
System (NAS) requires new or updated regulations, policies, and technologies to maintain safety and
enable efficient use of the airspace. One enabling technology to help address several UAS airspace
integration gaps are airspace encounter models, which have been fundamental to quantifying airborne
collision risk for manned and unmanned operation [
1
–
5
]. These models represent how aircraft behavior
and their relative geometries evolve during close encounters. They have supported the development
of surveillance and communication requirements [6,7].
1.1. Motivation
Mitigations for airborne collision risk and optimization of airspace operations are strongly dependent
on the distribution of geometries and behavior of aircraft encounters. For example, collision avoidance
systems are designed to determine, communicate, and coordinate avoidance maneuvers when they
determine that a maneuver is needed to avoid a collision. These systems are the last and third layer for
airspace conflict management and are employed after separation provision and strategic mitigation
have failed. Collision avoidance systems may leverage vehicle-to-everything (V2X) communication
technologies to improve performance and safety. How an aircraft behaves will influence the development
of V2X routing protocols, link budgets, and energy requirements [8].
Fast-time Monte Carlo simulations are often utilized to evaluate the performance of aviation
safety systems, such as detect and avoid (DAA) for UAS, for close encounters between aircrafts [
2
–
4
].
The design and effectiveness of these simulations are dependent on how close encounters are defined.
The relative geometries and separation between aircraft are important criteria when defining these
encounters. Simulated encounters are generated based on what safety function is being evaluated
and the performance or behavior of the aircraft involved. For example, encounters meant to prompt
Aerospace 2020, 7, 158; doi:10.3390/aerospace7110158 www.mdpi.com/journal/aerospace
