Assurance Methods for Increasingly Autonomous Systems
The inability to establish appropriate assurance for increasingly autonomous systems presents a critical barrier to their safe deployment in aviation and other safety-critical domains. This challenge drives development costs uneconomically high, delays adoption, and creates unknown and unmanageable risks. This talk explores the fundamental question: What constitutes sufficient evidence to substantiate safety claims
for increasingly autonomous systems performing safety-critical functions? In this talk we establish foundational concepts of safety assurance, distinguishing between design safety and operational safety, and examine why assurance remains fundamentally difficult. The core of the talk explores two complementary research thrusts. First, we investigate assurance of machine learning systems through a case study of AssistNet, a deep learning model for runway and taxiway detection. Using formal methods tools, we demonstrate how formal bound propagation and branch-and-bound algorithms can provide mathematical guarantees of neural network robustness against defined perturbations within an operational design domain. This work reveals both the promise and current limitations of formal verification for safety-critical computer vision applications. Second, we examine how machine learning can accelerate assurance processes themselves. We present machine-learning enhancements that can democratize formal theorem proving by translating natural language specifications into formal proofs, auto-suggesting proof tactics, and performing autonomous verification attempts. The presentation concludes by postulating paths forward to develop standardized, scalable certification pathways. While no single technique suffices, a combination of means, methods, tools and techniques can help in the assurance of increasingly autonomous aviation systems, thereby maintaining the rigorous safety standards the public demands while enabling innovation.
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
Loading virtual attendance info...
Speakers
Dr. Natasha Neogi
Biography:
Dr. Natasha Neogi is currently the Senior Technologist for Assured Intelligent Flight Systems at the National Aeronautics and Space Administration (NASA) and the former subproject manager of the System Wide Safety Project’s Safety Demonstrator Series. Her primary research interests are in the verification and validation of software-intensive safety-critical infrastructure systems, as well as certification issues concerning airworthiness of non-conventionally piloted vehicles. Previously, she was a staff scientist in the Office of the Chief Scientist, NASA Headquarters. She received her Ph.D. in Aeronautical and Astronautical Engineering from the Massachusetts Institute of Technology. She is a fellow of the AIAA and was the recipient of the AIAA Robert A. Mitcheltree and PEC Doug P. Ensor Young Engineer awards as well as NASA’s 2021 Outstanding Leadership Medal. She has numerous awards and publications in AIAA, IEEE and ACM conferences and journals.