Precision autonomous approach and landing using a novel vision and imitation learning method in GPS denied environment
The objective of the project is to develop a novel control methodology for autonomous approach and landing on the deck of a ship. This procedure is inspired by one that Navy pilots follow for landing on space-constrained ship decks, by utilizing a standardized visual reference called the “horizon bar”. The idea is to utilize machine vision to track a portable standardized visual cue, which a soldier can easily carry and install on a moving vehicle or a ship (similar to the “horizon bar”). Then combine the relative position and heading information from the vision system with imitation-learning-based control algorithms to bring an experienced pilot’s skill and decision-making process to a complex maneuver such as landing on a small ship deck or a dense urban terrain.
The video below shows the successful implementation of this autonomous tracking and landing method with a quadcopter:
A comprehensive helicopter flight dynamics code is developed based on the UH-60 helicopter and named Texas A&M University Rotorcraft Analysis Code (TRAC). This is a complete software packages, which could perform trim analysis to autonomous flight simulation and the capability to model any helicopter configuration. Different components of the helicopter such as the main rotor, tail rotor, fuselage, vertical tail, and horizontal tail are modeled individually as different modules in the code and integrated to develop a complete UH-60 model. Since the code is developed on a module basis, it can be easily modified to adopt another component or configure a different helicopter. TRAC can predict the dynamic responses of both the articulated rotor blades and the helicopter fuselage and yields the required pilot control inputs to achieve trim condition for different flight regimes such as hover, forward flight, coordinated turn, climb/descent, etc. These trim results are validated with the test data obtained from theUH-60 flight tests conducted by the US Army. Beyond trim analysis, TRAC can also generate linearized models at various flight conditions based on a first-order Taylor series expansion. The extracted linear models show realistic helicopter dynamic behavior and were used to simulate a fully autonomous flight that involves a UH-60 helicopter approaching a ship and landing on the deck by implementing a Linear Quadratic Regulator (LQR) optimal controller.
A visualization of this technology in action in a UH-60 Blackhawk helicopter
This project is being developed primarily by Bochan Lee. Bochan graduated from the Republic of Korea Naval Academy in 2010 and served as a Navy UH-60 helicopter pilot. In fall of 2016, he was selected by the South Korean government’s program to foster future specialists in aerospace engineering and joined the AVFL at Texas A&M University. Bochan has been working under the guidance of Dr. Moble Benedict with a focus on dynamics and control since 2016.