Flight Critical Software and Systems Development Using ASSERT

Kit Siu, Abha Moitra, Michael Durling, Meng Li, Han Yu, Heber Herencia-Zapana, Mauricio Castillo-Effen, Shiraj Sen, Craig Mc Millan, Daniel Russell, Sundeep Roy and Panagiotis Manolios.
Digital Avionics Systems Conference, (DASC) 2017, 2015© IEEE
(Best of track, Best of session awards)

Abstract

The size and complexity associated with software that monitors, controls, and protects flight critical products continues to grow. This is compounded by an increased use of autonomous systems which are just as complex, if not more so, since many operator responsibilities are supported and replaced by software in unmanned systems. Further, these systems are subject to cyber-enabled attacks, thereby necessitating another level of complex software to ensure security. General Electric has devoted a team to research and develop a new suite of tools to address the challenges with design, development, and verification of these software-intensive products. The goals are to develop technology, processes, and tools that result in more efficient software and system development as measured by cost and cycle time, and to enable new capabilities such as autonomy and the Industrial Internet. This paper will introduce the GE approach to formal requirements capture, requirements analysis, and auto test generation. We will introduce the ASSERT tool chain (Analysis of Semantic Specifications and Efficient generation of Requirements-based Tests). We will demonstrate aspects of the tool on an autonomous aerial inspection system.