Approach Towards Semi-Automated Certification for Low Criticality ML-Enabled Airborne Applications
Chandrasekar Sridhar, Vyakhya Gupta, Prakhar Jain, Karthik Vaidhyanathan
TL;DR
The paper addresses certification gaps for ML-enabled, low-criticality airborne systems under DO-178C Level D by proposing a semi-automated process anchored on a tri-axis classifier $C=\langle c_{\text{crit}}, c_{\text{aut}}, c_{\text{model}}\rangle$ and an Assurance Profile producing a final score $S_{\text{cert}}$ and confidence level $\sigma$. It demonstrates the approach through the Air Sight case study (YOLOv8 object detection) across Development, V&V, QA, and CM, with emphasis on data quality, model validation, and risk management. Preliminary results show a Final Assurance of $74.7\%$ (Moderate) with strong V&V signals but gaps in QA/CM and a drift-monitoring recertification plan rather than full recertification, indicating practical viability for low-criticality MLS with ongoing evaluation. The work sets the stage for extending the framework to broader use cases and higher-criticality domains while refining post-certification update practices.
Abstract
As Machine Learning (ML) makes its way into aviation, ML enabled systems including low criticality systems require a reliable certification process to ensure safety and performance. Traditional standards, like DO 178C, which are used for critical software in aviation, do not fully cover the unique aspects of ML. This paper proposes a semi automated certification approach, specifically for low criticality ML systems, focusing on data and model validation, resilience assessment, and usability assurance while integrating manual and automated processes. Key aspects include structured classification to guide certification rigor on system attributes, an Assurance Profile that consolidates evaluation outcomes into a confidence measure the ML component, and methodologies for integrating human oversight into certification activities. Through a case study with a YOLOv8 based object detection system designed to classify military and civilian vehicles in real time for reconnaissance and surveillance aircraft, we show how this approach supports the certification of ML systems in low criticality airborne applications.