Human-Centered Design for Connected Automation: Predicting Pedestrian Crossing Intentions

TL;DR

This paper models pedestrian crossing intentions in the context of level-5 autonomous vehicles, emphasizing the importance of communication and human-centered design to improve safety and interaction.

Contribution

It extends the Theory of Planned Behavior to include safety, trust, and understanding, providing insights for designing effective human-machine interfaces in autonomous driving.

Findings

Perceived safety and understanding strongly influence crossing intentions.

Social information significantly impacts pedestrians' decision-making.

The model guides design of communication strategies for safer pedestrian-ADS interactions.

Abstract

More than half of the 1.19 million annual traffic fatalities globally involve vulnerable road users, such as pedestrians, with a significant proportion attributable to human error. Level-5 automated driving systems (ADSs) have the potential to reduce these incidents; However, their effectiveness depends not only on automation performance but also on their ability to communicate intent and coordinate safely with pedestrians in the absence of traditional driver cues. This study aims to model pedestrian decision-making in road-crossing scenarios involving level-5 ADSs by extending the Theory of Planned Behavior (TPB) with safety, trust, compatibility, and understanding. An online survey (n = 212) found that perceived behavioral control, attitude, and social information significantly influence pedestrians' crossing intentions, with perceived safety and understanding having the strongest effects on the TPB constructs. The results offer guidance for designing eHMIs and cooperative V2X communication strategies that promote safe pedestrian-ADS interactions and advance human-centered design for autonomous vehicles.