A Structured Approach to Trustworthy Autonomous/Cognitive Systems
This work addresses the problem of ensuring trustworthiness in safety-related autonomous systems for developers and regulators, but it appears incremental as it builds on existing standards.
The paper tackles the lack of a generally accepted approach to ensure trustworthiness in autonomous/cognitive systems using AI, and proposes a framework based on a reference lifecycle enhanced from current safety standards to address this gap.
Autonomous systems with cognitive features are on their way into the market. Within complex environments, they promise to implement complex and goal oriented behavior even in a safety related context. This behavior is based on a certain level of situational awareness (perception) and advanced de-cision making (deliberation). These systems in many cases are driven by artificial intelligence (e.g. neural networks). The problem with such complex systems and with using AI technology is that there is no generally accepted approach to ensure trustworthiness. This paper presents a framework to exactly fill this gap. It proposes a reference lifecycle as a structured approach that is based on current safety standards and enhanced to meet the requirements of autonomous/cog-nitive systems and trustworthiness.