Safety Systems Engineering in Autonomy

Safety systems engineering in autonomy is the discipline concerned with rigorously demonstrating that complex, software-intensive systems—autonomous vehicles, aircraft, medical devices—will behave safely even under failure conditions or unforeseen inputs. Central to the work are structured frameworks like ISO 26262 for automotive functional safety and processes such as Automotive SPICE, which provide systematic methods for identifying hazards, assigning risk levels, and tracing requirements through to verified implementations. A recurring challenge is building convincing assurance cases: structured arguments, backed by evidence, that regulators and developers can use to justify confidence in a system whose behavior may emerge from learned models rather than hand-written logic. Active research focuses on how to certify systems that incorporate machine learning components, where traditional assumptions about deterministic, fully specified software break down, and on how security threats—which can deliberately drive a system into unsafe states—should be integrated into safety analysis from the outset.

Works

26,631

Total citations

68,832

Keywords

Safety AssuranceAssurance CasesSoftware CertificationFunctional SafetyAutomotive SPICEModel-Based Development