Fully Autonomous Taxi, Take-Off and Landing Aircraft Control System Using Computer Vision for Localization and Collision Avoidance
Is this project an undergraduate, graduate, or faculty project?
Undergraduate
Project Type
individual
Campus
Daytona Beach
Authors' Class Standing
Nishant Sharma, Senior
Lead Presenter's Name
Nishant Sharma
Lead Presenter's College
DB College of Engineering
Faculty Mentor Name
None
Abstract
The objective of this research will be to design a control system to completely automate the different stages of flight (taxi, take-off, cruise, and landing) using computer vision for localization and collision avoidance. The computer vision technology will work using machine learning algorithms and trained using real flight data. The airplane will function much like a self-driving car, in that it will be aware of its surroundings and make flight maneuvers accordingly. This system will first be tested using Software-In-The-Loop (SIL) methods under controlled 3D simulation environments. This will be followed by Hardware-in-the-loop (HIL) testing of the control algorithms on a scaled down version of the final aircraft (a Cessna 172 for instance). The flight data acquired from the Embry-Riddle fleet of Cessna 172 aircraft could possibly be used to help train the machine learning programs used for localization. A remote-controlled override system will also be implemented where a remote pilot can take over the aircraft control in case of an emergency. The automation of these processes will help reduce the number of pilots needed inside a cockpit, increase fuel efficiency enhance air traffic management and improve safety.
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
Fully Autonomous Taxi, Take-Off and Landing Aircraft Control System Using Computer Vision for Localization and Collision Avoidance
The objective of this research will be to design a control system to completely automate the different stages of flight (taxi, take-off, cruise, and landing) using computer vision for localization and collision avoidance. The computer vision technology will work using machine learning algorithms and trained using real flight data. The airplane will function much like a self-driving car, in that it will be aware of its surroundings and make flight maneuvers accordingly. This system will first be tested using Software-In-The-Loop (SIL) methods under controlled 3D simulation environments. This will be followed by Hardware-in-the-loop (HIL) testing of the control algorithms on a scaled down version of the final aircraft (a Cessna 172 for instance). The flight data acquired from the Embry-Riddle fleet of Cessna 172 aircraft could possibly be used to help train the machine learning programs used for localization. A remote-controlled override system will also be implemented where a remote pilot can take over the aircraft control in case of an emergency. The automation of these processes will help reduce the number of pilots needed inside a cockpit, increase fuel efficiency enhance air traffic management and improve safety.