Improving AEBS Validation Through Objective Intervention Classification Leveraging the Prediction Divergence Principle

2025-07-10

AI Summary

In open-loop re-simulation (e.g., Field Operational Tests), objective discrimination between true-positive (TP) and false-positive (FP) AEBS activations remains challenging due to scenario parameter uncertainty and driver intervention—leading to bias in conventional subjective manual labeling. This paper proposes a regularized classification framework based on the Prediction Disagreement Principle (PDP), integrating a simplified AEBS model with open-loop re-simulation to enable fully automated, criterion-free TP/FP classification. The method significantly improves inter-rater consistency and decision transparency; validation confirms that its integration with human annotation enhances overall reliability. Our key contribution is the first application of PDP to AEBS verification, establishing a reproducible, objective, and interpretable automated assessment standard. This advances safety validation from subjective, experience-based practices toward a data-driven paradigm.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

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

Abstract

The safety validation of automatic emergency braking system (AEBS) requires accurately distinguishing between false positive (FP) and true positive (TP) system activations. While simulations allow straightforward differentiation by comparing scenarios with and without interventions, analyzing activations from open-loop resimulations - such as those from field operational testing (FOT) - is more complex. This complexity arises from scenario parameter uncertainty and the influence of driver interventions in the recorded data. Human labeling is frequently used to address these challenges, relying on subjective assessments of intervention necessity or situational criticality, potentially introducing biases and limitations. This work proposes a rule-based classification approach leveraging the Prediction Divergence Principle (PDP) to address those issues. Applied to a simplified AEBS, the proposed method reveals key strengths, limitations, and system requirements for effective implementation. The findings suggest that combining this approach with human labeling may enhance the transparency and consistency of classification, thereby improving the overall validation process. While the rule set for classification derived in this work adopts a conservative approach, the paper outlines future directions for refinement and broader applicability. Finally, this work highlights the potential of such methods to complement existing practices, paving the way for more reliable and reproducible AEBS validation frameworks.

Problem

• Distinguish false and true AEBS activations accurately
• Reduce biases in human-labeled intervention classification
• Enhance validation transparency with rule-based PDP approach

Innovation

• Rule-based classification using Prediction Divergence Principle
• Combines with human labeling for transparent classification
• Conservative approach with future refinement directions