Two -day Workshops on Monday & Tuesday - MT-1

ROBUST AND ADAPTIVE CONTROL WORKSHOP

Kevin A. Wise, The Boeing Company
Eugene Lavretsky, The Boeing Company
Naira Hovakimyan, Virginia Tech

8:30am - 5:00pm
Olympic

Abstract
The workshop is designed for engineers/scientists that have had a minimum of an undergraduate level control systems course (root locus, Bode, Nyquist, etc.) and have an understanding of using state space methods for modeling control systems. The attendees will learn methods in robust linear control design (optimal control), robust stability analysis, Lyapunov stability theory, model reference direct adaptive control, and its recently developed modifications. The workshop presents lessons learned in applying these methods to fighter aircraft and advanced weapon systems.

The workshop covers in detail the adaptive control design and analysis methods used on X-36 RESTORE, X-45A Unmanned Combat Air Vehicle (UCAV), and on JDAM MK-82 guided weapon, JDAM MK-84, and the Laser-JDAM. Day 1 of the workshop begins with items 1.0 through 4.0 below, and is a summary of theory and lessons learned in developing robust linear control algorithms. These sections summarize the authors experience from 20 years of applying robust optimal control methods to aircraft and missiles within the Boeing Phantom Works. These methods are used to develop the baseline control algorithms that then are augmented with adaptive control, starting with Section 5.0 where an overview of the Lyapunov stability theory and an introduction to the design and analysis of classical linear in parameters adaptive control systems are given. Next, approximation properties of artificial neural networks and their application to the design of direct adaptive systems are presented. Key design points are discussed and illustrated through various simulation examples, followed by overview of open problems and future research directions in adaptive control and its applications to flight control of unmanned aerial vehicles, (Section 14). Day 2 of the workshop (Sections 15 - 21) presents recent advances in adaptive control theory, known as "The Theory of Fast and Robust Adaptation" or, equivalently "L1 adaptive control design". Novel stable adaptive control architectures for various classes of uncertain systems are discussed and their closed-loop performance as well as stability/robustness margins are quantified analytically. Simulation examples are given to illustrate key features of the new L1 design paradigm.