Can Cars Gesture? A Case for Expressive Behavior Within Autonomous Vehicle and Pedestrian Interactions

Abstract

One of the major challenges that autonomous vehicles (AVs) face in an urban setting is communicating with other road users such as pedestrians. In this work, we investigated with what expressive behaviors we can endow AVs such that pedestrians readily recognize the underlying intent of the vehicles’ movements. The purpose of our study was to test the impact of expressive stopping behaviors on pedestrians’ decision to cross a road. We utilized a virtual reality (VR) environment in which participants would have to cross a street in the presence of an oncoming vehicle that may or may not stop. Next, we crafted several expressive AV behaviors conveying its intention to stop for the pedestrian. Then, for each expressive design we recorded how quickly a pedestrian determined that it was safe to cross the street. We also administered repeated surveys of their subjective experiences. Our findings suggest that expressive behaviors such as easing into a full stop or stopping farther away can help pedestrians make quicker decisions to cross the road. Additionally, stopping farther away from the pedestrian also resulted in higher subjective experience for sense of safety, confidence, and intention understanding. We propose further investigation into expressive behaviors such as easing into a stop and stopping farther away to convey yielding intentions to pedestrians in future work. <xref ref-type="fn" rid="fn1" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">¹</xref>