Towards a Systematic Risk Assessment of Deep Neural Network Limitations in Autonomous Driving Perception
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908987695300608 |
|---|---|
| author | Pavlitska, Svetlana Gerking, Christopher Zöllner, J. Marius |
| author_facet | Pavlitska, Svetlana Gerking, Christopher Zöllner, J. Marius |
| contents | Safety and security are essential for the admission and acceptance of automated and autonomous vehicles. Deep neural networks (DNNs) are widely used for perception and further components of the autonomous driving (AD) stack. However, they possess several limitations, including lack of generalization, efficiency, explainability, plausibility, and robustness. These insufficiencies can pose significant risks to autonomous driving systems. However, hazards, threats, and risks associated with DNN limitations in this domain have not been systematically studied so far. In this work, we propose a joint workflow for risk assessment combining the hazard analysis and risk assessment (HARA) following ISO 26262 and threat analysis and risk assessment (TARA) following the ISO/SAE 21434 to identify and analyze risks arising from inherent DNN limitations in AD perception. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_20895 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Towards a Systematic Risk Assessment of Deep Neural Network Limitations in Autonomous Driving Perception Pavlitska, Svetlana Gerking, Christopher Zöllner, J. Marius Cryptography and Security Computers and Society Machine Learning Safety and security are essential for the admission and acceptance of automated and autonomous vehicles. Deep neural networks (DNNs) are widely used for perception and further components of the autonomous driving (AD) stack. However, they possess several limitations, including lack of generalization, efficiency, explainability, plausibility, and robustness. These insufficiencies can pose significant risks to autonomous driving systems. However, hazards, threats, and risks associated with DNN limitations in this domain have not been systematically studied so far. In this work, we propose a joint workflow for risk assessment combining the hazard analysis and risk assessment (HARA) following ISO 26262 and threat analysis and risk assessment (TARA) following the ISO/SAE 21434 to identify and analyze risks arising from inherent DNN limitations in AD perception. |
| title | Towards a Systematic Risk Assessment of Deep Neural Network Limitations in Autonomous Driving Perception |
| topic | Cryptography and Security Computers and Society Machine Learning |
| url | https://arxiv.org/abs/2604.20895 |