Document
National Aeronautics and Space Administration
NASA Armstrong
Flight Research Center
Dynamics and Control Branch
Steve Jacobson
Chief, Dynamics and Controls
04 March 2015
Abstract
NASA Armstrong continues it’s legacy of exciting work in the area of
Dynamics and Control of advanced vehicle concepts. This
presentation describes Armstrong’s research in control of flexible
structures, peak seeking control and adaptive control in the Spring
of 2015.
Control of Flexible Structures: Supersonic aircraft
Integrated rigid / low frequency
structural dynamics modeling
for handling qualities, control
law evaluation and development
• Integrated flexible - rigid state space
matrices
– aero effects due to structural deformation – structural deformation due to aero Response at Cockpit excitation
• 6 - DOF simulation modeling
• Z - AERO generation of state - space
Normal Acceleration Response at CG
matrices
Dynamic structural models
Modeling Approach Stiff Wing Structural Dynamics
• Linear Structural Model Transfer Function, Symmetric WF4 Input to qb_gyro_200_dps Output • Mode Shapes Flight Test • NASTRAN Finite Element Model nDoF -20 -40 • Correction for state consistency 0 1 2 Magnitude, dB 10 10 10 • Unsteady Aerodynamic Model • ZAERO Panel Method • Transfer Function Approximation • Sensor and Actuator Dynamics -100 Phase, deg • Models have 260 Total States 0 1 2 10 10 10 0.5
Technical Challenges
Flexible Wing Instabilities
Coherence 10 20 30 40 50 60 70 80 90 100 • Piloted Simulation Frequency, Hz • Real - time • Numerical Stability • Verification & Validation • Too many parameters to identify • Which parameters are critical • Free Play modeling • System Uncertainty
The X - 56A Flexible Wing Flutter Suppression
Challenge
• Most of envelope is unstable
• Using a frequency based
MIMO control design method,
performance objectives were
met
• Accelerometers used in
control system to suppress
high frequency instabilities
Gamma Response to 1 Degree Command Takeoff Design Speed (high fuel) Design Speed (low fuel) Flight Sensors + Accelerometer feedback
Sim of Flutter Suppression using FOSS
Demonstrated Active Flutter Suppression on X - 56A models with simulated strain feedback from thousands of sensors from fiber optic shape sensors (FOSS) X - 56A - Active Flutter Suppression Simulation using FOSS FOSS Mapping (6 fibers) Body Freedom Flutter (BFF) Symmetric Wing Bending Torsion Flutter (SWBT) Strain Modes from NASTRAN Antisymmetric Wing Bending Torsion Flutter (AWBT)
Simulation of Virtual Deformation Control using
FOSS
X - 56A Virtual Deformation Control Using FOSS
FOSS can support a shape
tracking architecture
X - 56A Virtual Deformation Control Results Wing tip deformations
through virtual
Wing tip
deformation control or
deflection (mostly due to
direct strain/deformation
bending)
feedback
Modal coordinates Bending mode tracking X - 56A Virtual Deformation Control – Modal Reference Simulation Controller Airframe states Reference transformation Modal filter
Automated Cooperative Trajectories
Cooperative Trajectory (CT) Concept Assured Autonomy for Aviation Transformation ACT enables automated, distributed, multi - vehicle control.
Proactive, collaborative approach to separation assurance and wake turbulence avoidance.
Distributed Knowledge of Aircraft and Wake Locations Integration of ADS - B Messages with Autopilot Systems Two or more aircraft Continuous data - link communication Safe, Efficient Growth in Global Aviation (such as ADS - B Out/In) Operation as Meta - Aircraft using automated, multi - vehicle Parallel, closely - spaced trajectories with coordination for peer - to - peer separation assurance and wake avoidance.
reduced separation (0.5 – 2 NM) Reduced Airspace Congestion Automatic control to maintain separation Improved ATC Workload Probabalistic vortex models combined with real - time, in situ measurements to Ultra - Efficient Commercial Transports estimate the location of the wake Sustained, trimmed flight within the upwash portion of the lead The project goal is to demonstrate ACT using aircraft’s wake reduces the trailing aircraft’s total drag by up to 15%.
COTS technology (i.e. ADS - B datalink and Lower Cost per Mile modifications to existing autopilots) Reduced Particulate Emissions at Altitude 0.5 to 2 NM Separation* *not to scale
Hypersonics , Adaptive Control
Application of adaptive control to Differential Sideslip hypersonics , lifting bodies flap generates creates • Lateral Control Divergence Parameter opposing sideslip (LCDP) - > HTV - 2, s pace shuttle, X - 38, X - rolling moment Rolling moment • Insidious roll reversal nature of LCDP (stable roots, positive control derivative) initiated by flap deflection Adaptive controls • HTV - 2 6 - DOF simulation of LCDP event • Control methodologies – Retrospective Cost Adaptive Control - > Nominal Dr. Dennis Bernstein, U of Michigan Plant deg -2 Cmd Model • Sysense , Inc – Dr. Jason Speyer Run 1 nominal plant -4 Run 2 Flt 1 Roll Reversal – Modified Gain EKF to estimate system 0 1 2 3 4 5 – Extended LQG to solve for feedback LCDP controller Model 1.5 diff flap 0.5 0 1 2 3 4 5