Is this project an undergraduate, graduate, or faculty project?
Graduate
Project Type
individual
Campus
Daytona Beach
Authors' Class Standing
Zackrey Schraeder, Graduate student
Lead Presenter's Name
Zackrey Schraeder
Lead Presenter's College
DB College of Aviation
Faculty Mentor Name
Dr. Daniel Pleffken
Abstract
Uncrewed Aircraft Systems (UAS), or more commonly known as drones, have become increasingly used for a variety of applications, and more research has emerged studying the human-machine interaction and pilot performance metrics. Current research focuses on desired behaviors or aggregated pilot data which does not provide a holistic picture of an individual pilot’s flight behaviors, which are shaped by their own experience, preferences, and risk tolerance. The study investigates whether an individual pilot’s flight behaviors can be considered a pattern or if there is enough flight-to-flight variability to consider the behaviors inconsistent. A series of 20 flights were conducted in the LuGus Studios Liftoff: FPV Drone Racing simulator with the same course, aircraft and 3-lap objective. Telemetry data was logged from from the Xbox controller inputs into a .csv file with metrics under the control input, aircraft movement, and aircraft orientation domains. If the metrics had a coefficient of variability that fit within a certain threshold, the behaviors were considered a pattern. This research can be applied to several human factors and algorithm or machine use cases. The identification of pilot tendencies, desirable or not desirable, can be converted into a post-flight feedback tools to make improvements. Pilots can also be better paired with missions given their unique styles. The research can also be used for designing autonomous agents that fly missions with human demonstrations and better controller or aircraft design to accommodate how specific pilots fly. This study aims to provide a quantitative, multi-domain assessment of a single pilot’s flight behaviors which can be used to optimize either the human or machine element of this relationship.
Did this research project receive funding support (Spark, SURF, Research Abroad, Student Internal Grants, Collaborative, Climbing, or Ignite Grants) from the Office of Undergraduate Research?
No
Quantitative Analysis of a Single Operator’s Flight Behavior in Simulation
Uncrewed Aircraft Systems (UAS), or more commonly known as drones, have become increasingly used for a variety of applications, and more research has emerged studying the human-machine interaction and pilot performance metrics. Current research focuses on desired behaviors or aggregated pilot data which does not provide a holistic picture of an individual pilot’s flight behaviors, which are shaped by their own experience, preferences, and risk tolerance. The study investigates whether an individual pilot’s flight behaviors can be considered a pattern or if there is enough flight-to-flight variability to consider the behaviors inconsistent. A series of 20 flights were conducted in the LuGus Studios Liftoff: FPV Drone Racing simulator with the same course, aircraft and 3-lap objective. Telemetry data was logged from from the Xbox controller inputs into a .csv file with metrics under the control input, aircraft movement, and aircraft orientation domains. If the metrics had a coefficient of variability that fit within a certain threshold, the behaviors were considered a pattern. This research can be applied to several human factors and algorithm or machine use cases. The identification of pilot tendencies, desirable or not desirable, can be converted into a post-flight feedback tools to make improvements. Pilots can also be better paired with missions given their unique styles. The research can also be used for designing autonomous agents that fly missions with human demonstrations and better controller or aircraft design to accommodate how specific pilots fly. This study aims to provide a quantitative, multi-domain assessment of a single pilot’s flight behaviors which can be used to optimize either the human or machine element of this relationship.