Modified-Emergency Index (MEI): A Criticality Metric for Autonomous Driving in Lateral Conflict

TL;DR

This paper introduces the Modified-Emergency Index (MEI), a new criticality metric for autonomous driving that effectively quantifies lateral conflict risks, outperforming existing metrics in urban safety assessments.

Contribution

The paper proposes MEI, a refined criticality metric specifically designed for lateral conflicts in autonomous driving, validated on a large dataset, and shown to outperform existing metrics.

Findings

MEI outperforms ACT and PET in criticality quantification.

MEI accurately captures risk evolution in lateral conflicts.

Validation on Argoverse-2 dataset confirms MEI's effectiveness.

Abstract

Effective, reliable, and efficient evaluation of autonomous driving safety is essential to demonstrate its trustworthiness. Criticality metrics provide an objective means of assessing safety. However, as existing metrics primarily target longitudinal conflicts, accurately quantifying the risks of lateral conflicts - prevalent in urban settings - remains challenging. This paper proposes the Modified-Emergency Index (MEI), a metric designed to quantify evasive effort in lateral conflicts. Compared to the original Emergency Index (EI), MEI refines the estimation of the time available for evasive maneuvers, enabling more precise risk quantification. We validate MEI on a public lateral conflict dataset based on Argoverse-2, from which we extract over 1,500 high-quality AV conflict cases, including more than 500 critical events. MEI is then compared with the well-established ACT and the widely used PET metrics. Results show that MEI consistently outperforms them in accurately quantifying criticality and capturing risk evolution. Overall, these findings highlight MEI as a promising metric for evaluating urban conflicts and enhancing the safety assessment framework for autonomous driving. The open-source implementation is available at https://github.com/AutoChengh/MEI.