Model Driven Systems Development for Space Systems

Paper number

IAC-07-D1.3.04

Author

Dr. Bruce Chesley, Boeing Space and Intelligence Systems, United States

Coauthor

Mr. Michael Mott, United States

Coauthor

Dr. Lawrence Dale Thomas, National Aeronautics and Space Administration (NASA), United States

Coauthor

Coauthor

Mr. Shane Arnott, Australia

Coauthor

Mr. Erik Daehler, Boeing, United States

Year

2007

Abstract

Engineering a space system presents many unique challenges. Better solutions to these challenges will allow space systems to increase their impact on quality of life for humanity. Space systems promise to benefit quality of life through persistent environmental monitoring on a global scale, rapid expansion of scientific knowledge, and economic gains for larger numbers of countries and corporations around the world (cf., Cost Effective Earth Observation Missions, International Academy of Astronautics, October 2005). Unfortunately, uncertainty in development times and unpredictable cost overruns limit participation and delay these quality of life improvements. We believe that improved systems engineering practice reduces uncertainty in space systems development and speeds the delivery of benefit, both scientific and economic, to the greatest number of people.  To do so, we focus on one seemingly simple challenge with profound implications for end-to-end systems engineering: managing the disparate development demands of long-lead hardware and software. 

Hardware development and qualification (e.g., the satellite flight computer) follows a serial process which typically extends for years. Flight software development, on the other hand, is more successful following an iterative process. Software iterations are constructed and delivered before the (long-lead) flight computer is available; the time to integrate software with flight computer hardware and other components is short, and software updates will be delivered years after the satellite is in operation. The multi-disciplinary issues introduced by blending these distinct design styles are a major source of risk to program schedules and quality. 

This paper discusses an approach which applies advances in tools and methods for end-to-end systems engineering for delivering a complex space system within cost and schedule. We address these issues from a lifecycle perspective, beginning with architecture development through design realization and operations. In particular, we address the application of system modeling tools such as the Systems Modeling Language (sysML) and Unified Modeling Language (UML) and their application to produce and manage the development of engineering artifacts. We show how the use of the Rational Unified Process for Systems Engineering (RUP SE) explicitly addresses risk and reduces integration schedule uncertainty for systems that necessarily incorporate iterative and serial development processes. We present a case study for a notional satellite development program, FireSat, to introduce and exemplify the concepts.

Abstract document

IAC-07-D1.3.04.pdf

Manuscript document

IAC-07-D1.3.04.pdf (🔒 authorized access only).

To get the manuscript, please contact IAF Secretariat.