Using IDOS to Develop EAGLE into “Real” Flight Software to ...

offbeatnothingSoftware and s/w Development

Dec 2, 2013 (3 years and 7 months ago)

77 views

Using IDOS to Develop EAGLE into
“Real” Flight Software to

Support Responsive Missions

David Chen

Universal Space Lines

1501 Quail St., Suite 100

Newport Beach, CA 92660


2
nd

Responsive Space Conference

Apr. 19
-
22, 2004

Los Angeles, CA

2nd Responsive Space Conference

Overview


IDOS


Entry Guidance Problem


EAGLE Guidance Algorithm


Implementation in MAVERIC


Implementation in IDOS


Comparison


Results


Conclusion

2nd Responsive Space Conference

IDOS Objectives


Reduce Vehicle Development Cost


Develop efficient Flight Mechanics Analysis Tools for vehicle development


Develop efficient requirements definition and systems engineering processes


Develop efficient processes to develop and test advanced GN&C systems to a TRL of
6 or 7


Ensure Flight Mechanics technologies are available to all potential vehicle developers


Reduce Operational Cost


Develop GN&C systems that require little or no mission specific software updates or
pre
-
flight verification


Reduce effort for trajectory planning, GN&C system verification I
-
load generation and
sustaining flight software engineering for new missions or anomaly resolution


Contribute to Department of Defense Operational Needs


Advanced GN&C
enables

short notice “on demand” mission planning to adapt to new
missions without the need for extensive ground planning


2nd Responsive Space Conference

IDOS in the Life
-
cycle

Pre
-
Phase A

Advanced
Studies

Phase A

Prelim Analysis

Phase B

Definition

Phase C

Design

Phase D

Development

Phase E

Operate

Design &

Develop

Validate

Operate

IDOS

Workflow

IDOS

Products

GN&C capabilities assessment

Requirements flowdown

Mission optimization and DRMs

System evaluation through simulation

Advanced GN&C algorithms

Flight software

Monte Carlo Analysis

Collaborative reviews via Web

Real
-
time execution metrics

Avionics requirements

Mission plans

Mission iloads

Mission validation


IDOS is designed around 3 fundamental workflows


Some or all are useful during each phase of the life
-
cycle

IDOS is an Integrating Environment designed to support the flight
software life cycle needs of aerospace vehicles

2nd Responsive Space Conference

How does IDOS work?

z + 1

z

2

-

1

z

z
-

1

Flight Control Software

(FCSW) Development
System

Real Time Test
Environment (RTTE)

Web

IDOS Management

System (IMS)

Database

Web


A design environment
provides:


highly productive
development tools
(industry standard
Matlab/Simulink)


Standardized and
automated interfaces


A central, web
-
accessible database
manages common:


vehicle configuration data


simulation and validation
tools


mission planning and
optimization tools


Advanced GN&C
algorithms


A real
-
time test
environment provides:


Automatic generation of
prototype flight control
software from GN&C
developer’s models


EARLY identification of
software defects and
avionics requirements

2nd Responsive Space Conference

Entry Guidance Problem


Compute feasible entry trajectory


Boundary conditions on longitude and latitude & other state
variables (i.e., TAEM and entry interface)


Path constraints


dynamic pressure


heating rate


acceleration


Control constraints


max and min limits on angles of attack and bank


rate limits on angles of attack and bank


acceleration limits on angles of attack and bank


Track computed entry trajectory


Command angle of attack and bank based on reference
trajectory and measured errors


Nullify disturbances and uncertainties

2nd Responsive Space Conference

EAGLE Guidance Algorithm


Trajectory length sub
-
problem


Determine an appropriate trajectory length to satisfy
downrange:




Determine a drag profile according to trajectory
length:



Trajectory curvature sub
-
problem


Cross range angle:



Heading angle:

D
dE
dS
1





cos
cos
cos
D
dE
dR


D
dE
d



cos
sin


Dr
D
L
V
dE
d






tan
cos
cos
sin
cos
1
2


2nd Responsive Space Conference


Successive approximation approach

1.
Initialize trajectory length to great circle arc distance

2.
Solve trajectory length sub
-
problem

3.
Solve trajectory curvature sub
-
problem using drag
profile from previous step

4.
Revise trajectory length according to the downrange
error

5.
Repeat steps 2
-
4 until error tolerances are satisfied

EAGLE Guidance Algorithm

2nd Responsive Space Conference

EAGLE Guidance Algorithm


Feedback linearized controller


Drag tracking


Follows reference drag profile to achieve range requirements


Commands
L/D
cos



Heading tracking


Follows reference heading profile to achieve cross range
requirements


Commands
L/D
sin




Angle of attack modulation

2nd Responsive Space Conference

Implementation in MAVERIC


Written in C Code


Developed on UNIX platform


gcc


EMACS


dbx


EAGLE implementation


Euler integration


Bisection search for drag profile


Finite number of bank reversal points


Development time ~ 1 year

2nd Responsive Space Conference

Implementation in IDOS


Written in Matlab/Simulink


Accepts S
-
functions or Simulink blocks


Graphical output


Integrated debugger


Developed on a PC platform


EAGLE implementation


Runge
-
Kutta integration


Gradient search for drag profile


Continuous bank reversal points


Development time < 3 months

2nd Responsive Space Conference

Comparison


MAVERIC


Text output


Command line based
debugger


Non real
-
time
simulation


Supports C



IDOS


Graphical plots


GUI based debugger



Non real
-
time and
real
-
time simulation


Supports multiple
languages

2nd Responsive Space Conference

Results


X
-
33 suborbital entry






FCSW
-
DS


Capsule entry from orbit






CAV Entry

2nd Responsive Space Conference

Conclusion


IDOS capabilities demonstrated


Rapid implementation of EAGLE to IDOS


Integrated environment for design, development,
implementation, testing, and validation


Greater flexibility for implementing advance guidance
algorithms


EAGLE successfully tested in real
-
time


Brings new design and implementation issues to light


Not possible with MAVERIC