Pedestrian Behavior Prediction for Automated Driving: Requirements, Metrics, and Relevant Features

TL;DR

This paper analyzes the requirements for pedestrian behavior prediction in automated driving, proposing a new system-level metric evaluated on real-world data, emphasizing the importance of contextual cues for safety.

Contribution

It introduces a novel system-level metric for pedestrian prediction performance and highlights the significance of contextual cues through extensive evaluation.

Findings

The new metric effectively measures system-level prediction performance.

Contextual cues significantly improve pedestrian prediction accuracy.

System-level approach is crucial for safe automated driving.

Abstract

Automated vehicles require a comprehensive understanding of traffic situations to ensure safe and anticipatory driving. In this context, the prediction of pedestrians is particularly challenging as pedestrian behavior can be influenced by multiple factors. In this paper, we thoroughly analyze the requirements on pedestrian behavior prediction for automated driving via a system-level approach. To this end we investigate real-world pedestrian-vehicle interactions with human drivers. Based on human driving behavior we then derive appropriate reaction patterns of an automated vehicle and determine requirements for the prediction of pedestrians. This includes a novel metric tailored to measure prediction performance from a system-level perspective. The proposed metric is evaluated on a large-scale dataset comprising thousands of real-world pedestrian-vehicle interactions. We furthermore conduct an ablation study to evaluate the importance of different contextual cues and compare these results to ones obtained using established performance metrics for pedestrian prediction. Our results highlight the importance of a system-level approach to pedestrian behavior prediction.