A Machine Learning Perspective on Automated Driving Corner Cases

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Schmidt, Sebastian, Körner, Julius, Günnemann, Stephan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909840194928640
author Schmidt, Sebastian
Körner, Julius
Günnemann, Stephan
author_facet Schmidt, Sebastian
Körner, Julius
Günnemann, Stephan
contents For high-stakes applications, like autonomous driving, a safe operation is necessary to prevent harm, accidents, and failures. Traditionally, difficult scenarios have been categorized into corner cases and addressed individually. However, this example-based categorization is not scalable and lacks a data coverage perspective, neglecting the generalization to training data of machine learning models. In our work, we propose a novel machine learning approach that takes the underlying data distribution into account. Based on our novel perspective, we present a framework for effective corner case recognition for perception on individual samples. In our evaluation, we show that our approach (i) unifies existing scenario-based corner case taxonomies under a distributional perspective, (ii) achieves strong performance on corner case detection tasks across standard benchmarks for which we extend established out-of-distribution detection benchmarks, and (iii) enables analysis of combined corner cases via a newly introduced fog-augmented Lost & Found dataset. These results provide a principled basis for corner case recognition, underlining our manual specification-free definition.
format Preprint
id arxiv_https___arxiv_org_abs_2510_10653
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Machine Learning Perspective on Automated Driving Corner Cases
Schmidt, Sebastian
Körner, Julius
Günnemann, Stephan
Computer Vision and Pattern Recognition
For high-stakes applications, like autonomous driving, a safe operation is necessary to prevent harm, accidents, and failures. Traditionally, difficult scenarios have been categorized into corner cases and addressed individually. However, this example-based categorization is not scalable and lacks a data coverage perspective, neglecting the generalization to training data of machine learning models. In our work, we propose a novel machine learning approach that takes the underlying data distribution into account. Based on our novel perspective, we present a framework for effective corner case recognition for perception on individual samples. In our evaluation, we show that our approach (i) unifies existing scenario-based corner case taxonomies under a distributional perspective, (ii) achieves strong performance on corner case detection tasks across standard benchmarks for which we extend established out-of-distribution detection benchmarks, and (iii) enables analysis of combined corner cases via a newly introduced fog-augmented Lost & Found dataset. These results provide a principled basis for corner case recognition, underlining our manual specification-free definition.
title A Machine Learning Perspective on Automated Driving Corner Cases
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.10653