To define software product assurance methods for autonomous on-board software, where existing methods cannot be used or are difficult to be applied.

The presence of autonomous software systems as well as artificial intelligence components in onboard software is becoming increasingly widespread in space missions. As autonomous software is finding its way into more and more missions, having a unified approach for the product assurance of this type of software that can be made applicable across projects is essential.Many of the software product assurance requirements currently in place for space missions could be impacted by autonomous software, from fundamental topics like criticality analysis, down to detailed software quality metrics. It is likely that new statistical approaches for the verification and validation of autonomous software have to be put in place. In order to efficiently manage the Agency?s ambitious roadmap for future missions, it is important to agree on a common approach within the European space industry and disseminate best practices to ensure the most cost-effective approach for the quality assurance of this new technology.Through this activity, a strong focus is given on verification (e.g. the development of a quality model for autonomous software), as well as using results of past studies (SPAAS, HRAF, ...) and experiences to formulate requirements for the verification, validation and safety assurance of autonomous software. This activity can be seen as the natural evolution of the previous studies into practical requirements that can start to be applied in projects.The activity encompasses the following tasks:- Survey the current and planned use of on-board autonomous systems in current and upcoming European missions.- Define methods and tools to be used for the verification, validation and safety assurance of autonomous and intelligent on-board software, based on past studies and experiences, as well as state-of-the-art industry best practices. - Elaborate a software quality metrication model and a method for software criticality analysis for autonomous and intelligent on-board software.- Elaborate a preliminary set of example requirements to serve a input for the next revision of the ECSS-Q-ST-80C SW PA standard.