Description of Corner Cases in Automated Driving: Goals and Challenges

TL;DR

This paper discusses the importance of identifying and describing corner cases in automated driving to improve machine learning models and system robustness, highlighting the challenges in creating machine-interpretable descriptions.

Contribution

It provides an overview of the goals and challenges in developing machine-interpretable descriptions of corner cases in automated driving systems.

Findings

Limited CC data hampers ML model training.

Better understanding of CC can improve offline dataset analysis.

Enhanced descriptions can lead to improved online system performance.

Abstract

Scaling the distribution of automated vehicles requires handling various unexpected and possibly dangerous situations, termed corner cases (CC). Since many modules of automated driving systems are based on machine learning (ML), CC are an essential part of the data for their development. However, there is only a limited amount of CC data in large-scale data collections, which makes them challenging in the context of ML. With a better understanding of CC, offline applications, e.g., dataset analysis, and online methods, e.g., improved performance of automated driving systems, can be improved. While there are knowledge-based descriptions and taxonomies for CC, there is little research on machine-interpretable descriptions. In this extended abstract, we will give a brief overview of the challenges and goals of such a description.