A Comprehensive Safety Metric to Evaluate Perception in Autonomous Systems

TL;DR

This paper introduces a new safety metric for autonomous vehicle perception that considers object importance factors like velocity and potential collision damage, providing a comprehensive safety assessment score.

Contribution

The paper presents a novel safety metric that integrates multiple object parameters for a more accurate perception safety evaluation in autonomous systems.

Findings

The new metric effectively combines various object importance factors.

It outperforms existing metrics in real-world and virtual data evaluations.

The metric provides an easily interpretable safety score.

Abstract

Complete perception of the environment and its correct interpretation is crucial for autonomous vehicles. Object perception is the main component of automotive surround sensing. Various metrics already exist for the evaluation of object perception. However, objects can be of different importance depending on their velocity, orientation, distance, size, or the potential damage that could be caused by a collision due to a missed detection. Thus, these additional parameters have to be considered for safety evaluation. We propose a new safety metric that incorporates all these parameters and returns a single easily interpretable safety assessment score for object perception. This new metric is evaluated with both real world and virtual data sets and compared to state of the art metrics.