The ARINC 653 specification is an important block from the Integrated Modular Avionics (IMA) definition, and both emerged in the civil aviation sector to answer to problems also identified in the space world. The AIR — ARINC 653 Interface in RTEMS — innovation initiative stemmed from the interest of the European Space Agency (ESA) in the adoption of the ARINC 653 concept for space on-board software, and aimed at the utilization and re-utilization of components off the shelf (COTS), exploiting the use of the Real-Time Executive for Multiprocessor Systems (RTEMS), a free/opensource real-time operating system. However, the AIR activies went further ahead, and resulted, not only in the intended proof of concept of the utilization of ARINC 653 in space, but also in the definition of a general architecture for an ARINC 653-compliant RTOS, allowing the co-existence of different RTOS kernels in different partitions.
The "AIR-II: ARINC 653 Interface in Space RTOS - Industrial Initiative" activities continue the work done in AIR, with the goal of becoming closer to a real system by improving and completing the key ideas identified. Besides evolving the design of an ARINC 653-compliant RTOS, AIR-II strives to make prototyping more efficient, by uniformizing the integration of operating systems and using widely available production chain tools. The establishment of a uniform methodology for operating system integration benefits from lessons learned in the course of work with Linux, and opens room for the flexible integration of real-time and non-real-time operating systems.
Open design issues from AIR, which will be dealth with in AIR-II, concern: consolidation of robust spatial segregation features, including hardware-based memory protection mechanisms; enhancing timeliness attributes, by adding support for multiple mode-based schedules (defined as an additional service in ARINC 653, Part 2) and runtime process deadline violation monitoring; definition of a flexible and portable Application Executive (APEX) Interface; definition of space-specific functions, such as health monitoring functions for error processing at all the levels of the system.
The AIR-II consortium benefits from the experience of both FCUL and Skysoft researchers obtained from the results of the AMOBA (ARINC 653 simulator for modular space based applications) activity, and is sustained by a new partner, Thales Alenia Space, a key system integrator for ESA.