Concurrent Learning Model Reference Adaptive Control

Squeezing the most out of past and present information to improve the controller’s neural networks convergence speed

Rodney Rodríguez
Nov 23, 2020 · 11 min read
Image for post
Image for post
This look-down view of the X-36 Tailless Fighter Agility Research Aircraft on the ramp at NASA’s Dryden Flight Research Center, Edwards, California, clearly shows the unusual wing and canard design of the remotely-piloted aircraft. NASA and Boeing developed the Reconfigurable Control for Tailless Fighter Aircraft (RESTORE) flight control software as a demonstration of the adaptability of the Model Reference Adaptive Controller with a neural-net algorithm to compensate for in-flight damage or malfunction of effectors, such as flaps, ailerons, and rudders. Credits: NASA.

Adaptive Control — Mimicking Your Brain’s Learning Abilities

A Proper Learning Strategy is the Name of the Game

Image for post
Image for post
Image for post
Image for post
MIT rule + σ Rubustifying Modification (upper), MIT rule + ε Rubustifying Modification (middle), and MIT rule + Optimal Rubustifying Modification (bottom).
Image for post
Image for post

The Concurrent Learning Concept

Image for post
Image for post

The Wing-Rock Phenomenon

Image for post
Image for post
Visualization of the Von-Karman vortex street. This phenomenon is similar to that appearing in the aircraft’s forebody. Source Multi-phase Flow Science.
Image for post
Image for post
Smoke generators and yarn tufts are used for flow visualization studies on a F/A-18 flown by NASA’s Dryden Flight Research Center, Edwards, California, in its High Alpha Research Vehicle (HARV) program. The aircraft is at about 42 degrees angle of attack in this photo, taken with a wing-mounted camera. In this photo, the aircraft’s forebody asymmetric vortex can be clearly affecting the wing flowfield. Source NASA.
Image for post
Image for post

Standard .vs. Concurrent Learning Model Reference Adaptive Control

Image for post
Image for post
Image for post
Image for post
MATLAB code for the MIT + Sigma Robust Modification Learning Law vs the Concurrent Learning laws.

Simulation Results

Image for post
Image for post
Simulation results showing aircraft’s bank angle evolution with different controllers. Image by the author.
Image for post
Image for post
Simulation results showing the reference model dynamics matching error (bank angle error). Image by the author.
Image for post
Image for post
Simulation results showing the feedback-path ANN’s weight convergence. Image by the author.
Image for post
Image for post
Simulation results showing the feed-forward ANN’s weight convergence. Image by the author.

The Startup

Medium's largest active publication, followed by +754K people. Follow to join our community.

Rodney Rodríguez

Written by

Aerospace Engineer at Airbus Defence and Space with a great passion for Technology and Science.

The Startup

Medium's largest active publication, followed by +754K people. Follow to join our community.

Rodney Rodríguez

Written by

Aerospace Engineer at Airbus Defence and Space with a great passion for Technology and Science.

The Startup

Medium's largest active publication, followed by +754K people. Follow to join our community.

Medium is an open platform where 170 million readers come to find insightful and dynamic thinking. Here, expert and undiscovered voices alike dive into the heart of any topic and bring new ideas to the surface. Learn more

Follow the writers, publications, and topics that matter to you, and you’ll see them on your homepage and in your inbox. Explore

If you have a story to tell, knowledge to share, or a perspective to offer — welcome home. It’s easy and free to post your thinking on any topic. Write on Medium

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store