Towards a Systematic Risk Assessment of Deep Neural Network Limitations in Autonomous Driving Perception

2026-04-21

AI Summary

This study addresses the inherent limitations of deep neural networks (DNNs) in autonomous driving perception—particularly concerning generalization, efficiency, interpretability, plausibility, and robustness—which can introduce significant safety and security risks. Notably, a systematic framework for analyzing these multidimensional risks has been lacking. To bridge this gap, this work proposes a novel joint risk assessment workflow that integrates Hazard Analysis and Risk Assessment (HARA) from ISO 26262 with Threat Analysis and Risk Assessment (TARA) from ISO/SAE 21434. By structurally identifying and quantifying risks stemming from DNN limitations, the proposed approach provides the first comprehensive methodology for coordinated safety and security analysis in autonomous perception systems, thereby establishing a foundation for subsequent risk mitigation strategies.

Technology Category

• Natural Language Processing: Safety and Robustness
• Computer Vision: Vision for Robotics & Autonomous Driving
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Security and Privacy: Large-scale security measurements
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

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.

Problem

• Deep Neural Networks
• Autonomous Driving
• Perception
• Risk Assessment
• Safety

Innovation

• risk assessment
• deep neural networks
• autonomous driving perception
• HARA
• TARA