Practical Solutions for Machine Learning Safety in Autonomous Vehicles

TL;DR

This paper reviews practical machine learning safety techniques for autonomous vehicles, addressing safety, security, and user experience challenges to improve dependability of ML-based automotive software.

Contribution

It provides an organized mapping of safety strategies to current machine learning techniques, filling a gap in automotive safety standards for ML systems.

Findings

Mapped safety strategies to ML techniques for autonomous vehicles

Identified security limitations of ML components

Discussed user experience considerations in ML safety

Abstract

Autonomous vehicles rely on machine learning to solve challenging tasks in perception and motion planning. However, automotive software safety standards have not fully evolved to address the challenges of machine learning safety such as interpretability, verification, and performance limitations. In this paper, we review and organize practical machine learning safety techniques that can complement engineering safety for machine learning based software in autonomous vehicles. Our organization maps safety strategies to state-of-the-art machine learning techniques in order to enhance dependability and safety of machine learning algorithms. We also discuss security limitations and user experience aspects of machine learning components in autonomous vehicles.