Automatic Flight Control System Design for an Unmanned Research Vehicle Using Discrete Quantitative Feedback Theory
Abstract
This thesis presents the application of non-minimum phase (NMP) w'- plane discrete MIMO (multiple-input-multiple-output) Quantitative Feedback Theory (QFT) to the design of a three-axis rate-commanded automatic flight control system for an unmanned research vehicle (URV). The URV model used has seven inputs and three outputs derived from the small-angle perturbation equations of motion. Plant parameter uncertainty consists of six flight conditions derived from variations in the aircraft center of gravity, airspeed, and gross weight. A weighting matrix delts is used to post-multiply the plants for blending the seven effector inputs into three effective rate-command inputs and resulting in an effective plant Pe = P recursive. Second order effector models and first order feedback sensor models are included in the plant. A time- scaled recursive algorithm is used to transform the continuous plant models to the w'-plane thereby avoiding the numeric problems associated with an intermediate z-plane representation. All the URV plant elements are minimum phase (MP). The transformation produces, however, a sampling NMP zero and one other NMP zero due to the three pole excess actuator/sensor model elements. These NMP elements limit the available loop bandwidth (li(jv)). Standard QFT design is used, with plant templates P = (P(jv)) which quantitatively express the plant uncertainty. Due to the loop bandwidth limitations, only stability bounds are derived. The loop transmissions (li(jv)) are then shaped to achieve the maximum levels subject to the stability bounds. This is followed in the usual QFT manner with design of prefilters.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1990
- Accession Number
- ADA230364
Entities
People
- David G. Wheaton
Organizations
- Air Force Institute of Technology