Machine Learning Revolutionizes Navy Helicopters Landing on Ships

This post is also available in: עברית (Hebrew)

One of the most difficult tasks for Navy pilots is landing a helicopter on a ship. Those involved have to be incredibly coordinated and account for the ever-changing natural conditions, and automated solutions have come up short, until now.

A new project spearheaded by the US Navy with the involvement of researchers from Texas A&M University aims to revolutionize autonomous vertical takeoff and landing (VTOL) aircraft technology, trying to find a new way to land helicopters on ships without human intervention.

The researchers are reportedly working on an approach combining optimal aircraft design with a trained machine-learning algorithm to enable the fully autonomous landing of an aircraft.

Dr. Moble Benedict, associate professor in the Department of Aerospace Engineering at Texas A&M, explained that helicopter pilots do not focus on the moving deck when landing, but rather keep their eyes on a specialized piece of equipment known as the ‘horizon bar’, a gyro-stabilized lighted strip that provides an artificial horizon, acting as a point of reference for the pilot.

Instead of focusing on using cameras, GPS, and lidar like in previous studies, Benedict and his co-PI Dr. Dileep Kalathil decided to take a different route and aim to automate the entire landing process by replicating a pilot’s behavior while focusing on the horizon bar.

When asked what makes their approach truly innovative, Dr. Kalathil, assistant professor in the Department of Electrical and Computer Engineering said- “Reinforcement learning.” According to Interesting Engineering, this type of machine learning is the building block for the control algorithm directing the autonomous systems.

Furthermore, this fusion of aerospace engineering and electrical and computer engineering isn’t just theoretical since Benedict and Kalathil have already demonstrated successful tracking and safe landings of unmanned aerial systems (UAS) in various challenging conditions.

Benedict is designing VTOL aircraft that are gust-tolerant and efficient and may potentially feature foldable wings for a seamless transition from vertical takeoffs to fixed-wing cruising. The designs will be tested through simulations, wind tunnel analyses, and real-world flight tests.

Kalathil is focusing on algorithmic solutions, saying that “the real challenge lies in the unpredictable behavior of rough seas,” and revealing plans to use wind sensors for real-time adjustments, which would allow to counteract specific conditions, bridging the simulation-to-reality gap that has stymied previous efforts.

The Navy is looking for an aircraft that’s runway-independent, fuel-efficient for extended flights, and capable of safe, automated landings on a moving ship, and this project could very well be the answer.