Toward a Framework for Modeling Space Systems Architectures

Peter Shames, Joseph Skipper
2006 SpaceOps 2006 Conference   unpublished
There recently have been a number significant effort to develop powerful and extensible approaches for describing a general class of software intensive system architectures. These approaches are typically focused upon architectures of terrestrial systems and they range from generally applicable, but formalized, approaches like the System Engineering Modeling Language (SysML), to more focused approaches such as the Unified Modeling Language (UML) for RM-ODP (Reference Model for Open Distributed
more » ... r Open Distributed Processing). In alignment with the recommendations made in the IEEE 1471-2000 Recommended Practice for Architectural Description of Software-Intensive Systems, both of these approaches intend to support an appropriate set of viewpoints for developing system architectural descriptions. All of these standard architectural approaches are intended to describe large-scale terrestrial data systems that are inherently complex, but are typically fixed in one place and often designed and built by a single organization. In the world of space systems there is an even higher level of complexity in that these are most often multi-organizational creations that are best characterized as systems of systems. In further contrast to terrestrial systems the most challenging elements of these systems, the spacecraft, are not fixed in place, but are flying through space at high velocity, must use specialized ground and space communications assets and protocols, are often at great distance from the Earth and are frequently out of contact with their control centers. These attributes of space communications systems drive architectural complexity and require consideration of issues that are not typical in terrestrial systems. The set of viewpoints that the existing standard approaches define are not completely adequate to the task of describing space systems. Work has been done during the last couple of years to model space data systems using a methodology called the Reference Architecture for Space Data Systems (RASDS) that is derived from RM-ODP. This modeling approach augments the RM-ODP viewpoints on a system, that include the organizational (Enterprise), the abstract (informational, computational), and the more concrete (Engineering, Technology), by adding ones that deal directly with space communications and protocols, physical element connectivity, and their interactions with the environment. The efficacy of RASDS has been demonstrated by its use in several NASA projects to describe the end to end architectures of their spacecraft data systems. Recent work has been done at JPL in an internal research and development project called Model Based Engineering and Design (RAMSS) to extend this RASDS approach to capture all of the other physical aspects of space systems in an extended model. Modeling the design of space systems necessitates inclusion of one or more viewpoints that deal with the rest of the physical attributes of these systems and their interaction with the environment. These other attributes include mass, power, propulsion, thermal, structure and dynamic control, in addition to the gravitational and environmental aspects already in the RASDS connectivity viewpoint. This conceptual approach is intended to be general enough to permit description of civilian, military, and commercial space systems, the spacecraft physical and logical design, ground
doi:10.2514/6.2006-5581 fatcat:lmvuw4oaqncfpftzv2hbawikve