Situation Coverage Based Safety Analysis of an Autonomous Aerial Drone in a Mine Environment

Abstract—If we want to integrate autonomous aerial drones into safety-critical contexts, particularly in dynamic and hazardous environments like mining operations, we need to rigorously assure their safety. Despite significant technological advancements in drone technology over the past decade, this remains a challenge. The current safety engineering methods employed in drones cannot demonstrate convincingly that AI techniques can effectively mitigate unsafe situations with a specified level of confidence and reliability. In this paper, we present a brief study of various approaches, with particular focus on the situation coverage-based approach. A key challenge lies in identifying a finite set of representative situations for testing from the infinite possibilities that could occur in real-world scenarios. This research contributes to advancing our understanding of situation coverage based safety assessment methodologies and coverage criteria.