Qualification of Proof Assistants, Checkers, and Generators: Where Are We and What Next?

Cyber-physical systems, such as learning robots and other autonomous systems, employ high-integrity software in their safety-critical control. This software is developed using a range of tools some of which need to be qualified for this purpose according to international standards. In this article, we first evaluate the state of the art of tool qualification for proof assistants, checkers (e.g., model checkers), and generators (e.g., code generators, compilers) by means of a SWOT (Strengths, Weaknesses, Opportunities, Threats) analysis. Our focus is on the qualification of tools in the three mentioned categories. Our objective is to assess under which conditions these tools are already fit or could be made fit for use in the practical engineering and assurance of high-integrity control software. In a second step, we derive a viewpoint and a corresponding range of suggestions for improved tool qualification from the results of our SWOT analysis.