Join on Zoom: https://iastate.zoom.us/j/99516192680 

Defeater-Driven Framework for Software Safety Assurance in sUAS

The rapid growth of small Uncrewed Aircraft Systems (sUAS) across diverse applications (e.g., agriculture, package delivery) has resulted in a corresponding dramatic increase in the number of reported incidents. This also introduces operational complexities within shared airspaces and increases concern for the safety of the public and property, calling for a robust and transparent framework to mitigate risks. Safety Assurance Cases, which present structured arguments supported by evidence, are a widely used approach to demonstrate safety claims in safety-critical systems. However, assurance cases have been traditionally applied to more static and deterministic use cases, and do not fully capture the contextual, socio-technical, or operational factors that may invalidate or weaken key assumptions. These argument-undermining factors, known as defeaters, pose a significant challenge to building trustworthy safety justifications for sUAS. This work presents a defeater-driven framework for safety assurance in sUAS. First, we explore what an sUAS system looks like, including the elicitation of requirements and the discovery of operational risks. Second, we propose a general taxonomy of real-world defeaters to improve defeater identification and coverage. Third, we develop a human-in-the-loop approach to automate the process of identifying defeaters within assurance cases. Integrating these insights, the framework seeks to provide systematic mechanisms for eliciting, organizing, and analyzing defeaters throughout the assurance case lifecycle, ultimately strengthening the transparency, robustness, and credibility of safety arguments for increasingly complex sUAS operations.

Committee: Robyn Lutz (major professor), Myra Cohen, Simanta Mitra, Samik Basu, and Xiaoqiu Huang