Implementation of Road Safety Perception in Autonomous Vehicles in a Lane Change Scenario

TL;DR

This paper explores how autonomous vehicles can incorporate human safety perceptions into lane change scenarios, highlighting differences between subjective safety perceptions and objective safety metrics.

Contribution

It introduces a framework for integrating human safety perception into autonomous vehicle algorithms, specifically addressing lane change scenarios.

Findings

Significant differences between perceived and objective safety metrics.

Perception varies with maneuver type and driver location.

Highlights need for aligning safety measures with human perceptions.

Abstract

Understanding human driving behavior is crucial to develop autonomous vehicles' algorithms. However, most low level automation, such as the one in advanced driving assistance systems (ADAS), is based on objective safety measures, which are not always aligned with what the drivers perceive as safe and their correspondent driving behavior. Finding the bridge between the subjective perception and objective safety measures has been analyzed in this paper focusing specifically on lane-change scenarios. Results showed statistically significant differences between what is perceived as safe by drivers and objective metrics depending on the specific maneuver and location of drivers.