Combining Evasive and Braking Reactions for Safety Reference Models in Automated Vehicles

2026-09-24

AI Summary

This study addresses the limitation of existing safety reference models that predominantly focus on longitudinal braking while neglecting lateral steering for collision avoidance. We propose a hybrid fuzzy safety model that, for the first time, integrates the Potential Field Safety (PFS) and Critical Fuzzy Safety with Lane Change (CFS-LC) metrics into a finite-state architecture incorporating perception-reaction delays. By leveraging fuzzy logic, this framework unifies the modeling of human drivers’ cooperative braking and steering decision-making processes. The proposed model effectively expands the scope of preventable accidents while preserving behavioral plausibility and computational efficiency. Ultimately, it provides a transparent and interpretable human reference model suitable for the benchmarking of autonomous driving systems.

Abstract

Computational models of careful and competent human drivers are essential for scenario-based evaluation of automated driving systems (ADS). However, most existing safety reference models primarily focus on longitudinal braking, neglecting the role of evasive steering in human collision avoidance. This paper proposes a hybrid Fuzzy-Safety Model (FSM-H) that integrates longitudinal mitigation and lateral avoidance within a unified behavioral framework. The braking component is governed by Proactive Fuzzy Safety (PFS) metrics, representing the erosion of longitudinal safety margins, while the steering component is driven by Criticality Fuzzy Safety for lane-change (CFS-LC), capturing lateral conflict severity and maneuver feasibility. A finite-state architecture models the sequential escalation from nominal driving to braking and, when necessary, to evasive steering, incorporating perception-reaction time and lane-check delays to reflect human decision processes. The model is evaluated in reconstructed high-criticality cut-in scenarios and compared with braking-only and steering-only reference strategies. Results show that the hybrid approach expands the preventability envelope while maintaining behavioral plausibility and computational tractability. The proposed framework provides a transparent and explainable human reference model suitable for simulation-based ADS safety benchmarking and regulatory assessment.

Problem

• Automated Vehicles
• Safety Reference Models
• Collision Avoidance
• Evasive Steering
• Scenario-based Evaluation

Innovation

• Fuzzy-Safety Model
• Evasive Steering
• Longitudinal Braking
• Finite-State Architecture
• Automated Driving Systems