L1 adaptive control flight testing and extension to nonlinear reference systems with unmatched uncertainty

Ackerman, Kasey Alan

Permalink

https://hdl.handle.net/2142/113855 Copy

Description

Title

L1 adaptive control flight testing and extension to nonlinear reference systems with unmatched uncertainty

Author(s)

Ackerman, Kasey Alan

Issue Date

2021-12-01

Director of Research (if dissertation) or Advisor (if thesis)

Hovakimyan, Naira

Doctoral Committee Chair(s)

Hovakimyan, Naira

Committee Member(s)

• Voulgaris, Petros
• Salapaka, Srinivasa
• Stipanovic, Dusan
• Theodorou, Evangelos
• Gregory, Irene

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Adaptive Control
• L1
• Flight Testing
• Nonlinear Systems

Abstract

Building upon prior research efforts deploying L1 adaptive control in remotely piloted aerospace applications, this dissertation presents the progression of in-flight evaluation of L1 adaptive control to manned flight testing on Calspan’s variable stability Learjet and to an augmentation of an autonomous trajectory planner on a multirotor aircraft. These efforts ultimately led to the development of a new L1 adaptive controller for a class of control-affine nonlinear reference systems subject to time-varying, state-dependent matched and unmatched uncertainties. The L1 adaptive controller for the Learjet flight tests was designed as stability augmentation system, modifying the pilot's stick-to-surface commands, and was evaluated in a series of flying and handling qualities tests. The results of the Learjet flight tests demonstrated the ability of the L1 adaptive controller to recover desired flying qualities and safe, consistent handling qualities in the presence of off-nominal dynamics, some of which had severe flying qualities deficiencies and aggressive tendencies toward adverse pilot-aircraft interaction, and simulated aircraft failures. A modification of the Learjet control law was implemented, with a nonlinear reference system and estimation of both matched and unmatched uncertainties, for a multirotor aircraft as an augmentation of a geometric trajectory-tracking baseline controller, tracking a reference trajectory generated by a model predictive path integral trajectory planner. Simulation results demonstrated that, with the L1 augmentation, the vehicle was able to navigate a complex environment in the presence of uncertainty and external disturbances. The new L1 adaptive controller provides a theoretical foundation for the L1 augmentation in the multirotor application, and may be applicable to tilt-rotor, tilt-wing, and split-propulsion vertical takeoff and landing aircraft proliferating in the urban air mobility sector. The theory is based on incremental stability for robust trajectory tracking and uses a piecewise-constant adaptive law. It proposes a feedforward compensator (in the form of an embedded linear parameter-varying system), synthesized for the variational dynamics of the system using linear matrix inequality-based robust control methods to minimize the peak-to-peak gain from unmatched uncertainty to the system state. A realization of the feedforward compensator in the ambient space can be directly applied to the nonlinear system. Analysis of the closed-loop system provides an incremental stability guarantee and bounds the transient and steady-state trajectory-tracking error.

Graduation Semester

2021-12

Type of Resource

Thesis

Permalink

http://hdl.handle.net/2142/113855

Copyright and License Information

Copyright 2021 by Kasey Alan Ackerman. All rights reserved.

Owning Collections