Description of Corner Cases in Automated Driving: Goals and Challenges

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.

[1]  Yingda Xia,et al.  Synthesize then Compare: Detecting Failures and Anomalies for Semantic Segmentation , 2020, ECCV.

[2]  Florian Kuhnt.  Holistic Temporal Situation Interpretation for Traffic Participant Prediction , 2020 .

[3]  Franz Wotawa,et al.  Using Ontologies for Test Suites Generation for Automated and Autonomous Driving Functions , 2018, 2018 IEEE International Symposium on Software Reliability Engineering Workshops (ISSREW).

[4]  Germán Ros,et al.  CARLA: An Open Urban Driving Simulator , 2017, CoRL.

[5]  Petter Nilsson,et al.  Using Control Synthesis to Generate Corner Cases: A Case Study on Autonomous Driving , 2018, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[6]  Sebastian Ramos,et al.  The Cityscapes Dataset for Semantic Urban Scene Understanding , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Tejaswi Koduri,et al.  AUREATE: An Augmented Reality Test Environment for Realistic Simulations , 2018 .

[8]  Fabian Schuldt,et al.  Defining and Substantiating the Terms Scene, Situation, and Scenario for Automated Driving , 2015, 2015 IEEE 18th International Conference on Intelligent Transportation Systems.

[9]  Bernhard Sick,et al.  Out-of-distribution Detection and Generation using Soft Brownian Offset Sampling and Autoencoders , 2021, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[10]  Lutz Eckstein,et al.  A framework for definition of logical scenarios for safety assurance of automated driving , 2019, Traffic injury prevention.

[11]  Tim Fingscheidt,et al.  Towards Corner Case Detection for Autonomous Driving , 2019, 2019 IEEE Intelligent Vehicles Symposium (IV).

[12]  Markus Maurer,et al.  A Deep-Learning Approach for the Detection of Overexposure in Automotive Camera Images , 2018, 2018 21st International Conference on Intelligent Transportation Systems (ITSC).

[13]  Ersin Yumer,et al.  Diverse Complexity Measures for Dataset Curation in Self-Driving , 2021, 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[14]  Michael Färber,et al.  AWARE: A Situational Awareness Framework for Facilitating Adaptive Behavior of Autonomous Vehicles in Manufacturing , 2020, SEMWEB.

[15]  Markus Maurer,et al.  Scenarios for Development, Test and Validation of Automated Vehicles , 2018, 2018 IEEE Intelligent Vehicles Symposium (IV).

[16]  Roland Siegwart,et al.  Fishyscapes: A Benchmark for Safe Semantic Segmentation in Autonomous Driving , 2019, 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW).

[17]  A. Pretschner,et al.  Tests für automatisierte und autonome Fahrsysteme , 2021, Informatik Spektrum.

[18]  Philip Koopman,et al.  How Many Operational Design Domains, Objects, and Events? , 2019, SafeAI@AAAI.

[19]  Senthil Yogamani,et al.  Weather and Light Level Classification for Autonomous Driving: Dataset, Baseline and Active Learning , 2021, 2021 IEEE International Intelligent Transportation Systems Conference (ITSC).

[20]  Marc René Zofka,et al.  Reliving the Dataset: Combining the Visualization of Road Users' Interactions with Scenario Reconstruction in Virtual Reality , 2021, ArXiv.

[21]  Pascal Fua,et al.  SegmentMeIfYouCan: A Benchmark for Anomaly Segmentation , 2021, NeurIPS Datasets and Benchmarks.

[22]  Junfeng Yang,et al.  DeepXplore: Automated Whitebox Testing of Deep Learning Systems , 2017, SOSP.

[23]  Martin Herrmann,et al.  Leveraging combinatorial testing for safety-critical computer vision datasets , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[24]  Tim Fingscheidt,et al.  Systematization of Corner Cases for Visual Perception in Automated Driving , 2020, 2020 IEEE Intelligent Vehicles Symposium (IV).

[25]  Tim Fingscheidt,et al.  An Application-Driven Conceptualization of Corner Cases for Perception in Highly Automated Driving , 2021, 2021 IEEE Intelligent Vehicles Symposium (IV).

[26]  Matthias Bitzer,et al.  System Co-Design (SCODE): methodology for the analysis of hybrid systems , 2020, Autom..

[27]  Georgios Fainekos,et al.  Requirements-Driven Test Generation for Autonomous Vehicles With Machine Learning Components , 2019, IEEE Transactions on Intelligent Vehicles.